Objective:

To examine the effect of orlistat on dietary restraint, disinhibition, hunger, and binge eating and to understand the relation between changes in eating behavior and weight maintenance.

Methods and Procedures:

Subjects were 306 women and men (age: 19–45 years; BMI: 37.5 4.1 kg/m²) included in the Scandinavian Multicenter study of Obese subjects with the Metabolic Syndrome, a 3-year clinical trial of orlistat or placebo following an 8-week very low energy diet (VLED). Outcomes were changes in weight and in the Three Factor Eating Questionnaire (TFEQ) and Binge Eating Scale (BES) between screening and 17 and 33 months after randomization. As reported previously, weight gain following VLED was lower in subjects treated with orlistat than with placebo.

Results:

Compared to screening results, dietary restraint was increased and disinhibition, hunger, and binge eating were decreased in both groups. These changes were similar in both groups with the exception of the hunger score at month 33 that was reduced more in the placebo than in the orlistat group (difference between groups -1.1 (95% CI (-2.0, -0.2)) P = 0.014). In multivariate analyses, scores for restraint, disinhibition and binge eating were associated with weight loss after adjustment for BMI, gender, age, and treatment (all P 0.002, model R ² = 0.12–0.17).

Discussion:

Orlistat did not affect eating behavior differently in any substantial way than the placebo did in this long-term weight maintenance trial. The results indicate that increased restraint and decreased disinhibition and binge eating are important for sustained weight maintenance in obese subjects with the metabolic syndrome.

Introduction

The relation between obesity and several of its comorbidities such as type 2 diabetes and premature atherosclerosis seems to be mediated by risk factors characterized by the metabolic syndrome (1,2). The metabolic syndrome is closely associated with abdominal obesity (3,4). Weight loss achieved by changes in eating behavior, diet and physical activity are the cornerstones in the treatment of the metabolic syndrome. However, weight control methods often produce short-term success, while sustained weight maintenance is difficult to achieve (5,6). Thus, pharmacological therapy has been proposed as an adjunct to diet and lifestyle changes to improve the maintenance of weight loss. Orlistat is a pancreatic lipase inhibitor that reduces the intestinal absorption of fat (7,8). Because gastrointestinal symptoms may occur after the intake of fatty food, orlistat is thought to act as a policing influence in daily food choices.

Psychological and behavioral factors including dietary restraint, disinhibition, hunger, and binge eating may also influence dietary intake and body weight. Dietary restraint is defined as the tendency to consciously restrict food intake either to prevent weight gain or to promote weight loss by control over energy intake and types of food eaten (9). Disinhibition is the tendency to overeat in the presence of palatable foods or other disinhibitors such as emotional distress (10). Hunger has been defined as the susceptibility to perceived body symptoms that signal the need for food (10). The Three Factor Eating Questionnaire (TFEQ) (11) and the Binge Eating Scale (BES) (12) are recognized instruments for assessing these eating behaviors.

Increasing dietary restraint, and decreasing disinhibition and hunger are important for improved weight maintenance (13,14,15,16,17). A better understanding of how eating behaviors influence weight maintenance during treatment with orlistat, a drug that may influence daily food choices, could be helpful in the clinical management of obesity.

We recently reported the results of the Scandinavian Multicenter study of Obese subjects with the Metabolic Syndrome, a placebo-controlled clinical trial aimed at investigating the efficacy of orlistat in the maintenance of weight loss following a very low energy diet (VLED) in 309 abdominally obese women and men. The VLED initiated a weight loss of 14.4 kg. After 18 months the maintained weight loss was 11.7 kg in the orlistat group and 9.6 kg in the placebo group and after 3 years the maintained weight loss was 9.4 kg and 7.2 kg respectively, in the orlistat and placebo groups (all P < 0.05). Thus, the gain in body weight after VLED at month 36 was less in the orlistat group compared to the placebo group (4.6 kg vs. 7.0 kg, P < 0.02) (18).

The objective of this study was to examine the effect of orlistat on dietary restraint, disinhibition, hunger and binge eating at month 17 and month 33 after an initial weight loss and to understand the relation between eating behavior and weight maintenance in subjects participating in the Scandinavian Multicenter study of Obese subjects with the Metabolic Syndrome trial.

Methods and Procedures

Subjects

Subjects were recruited at nine clinical research centers in Scandinavia (Denmark, Finland, Norway, and Sweden) to participate in the Scandinavian Multicenter study of Obese subjects with the Metabolic Syndrome as described elsewhere (18). Men and women were eligible for the trial if they met the following criteria: (i) abdominal obesity (waist circumference >92 cm for females and >102 cm for males) (ii) at least one of the following metabolic risk factors: impaired fasting glucose (>6.1 mmol/l and <7.1 mmol/l), diet treated type II diabetes, dyslipidemia (HDL cholesterol <1.1 mmol/l for females and 0.9 mmol/l for males) and/or triglycerides >2.0 mmol/l. Subjects were excluded if they had clinically relevant conditions that might affect the outcome of the trial (i.e., weight change of more than 4% for the previous 2 months before inclusion, women of child-bearing potential who were not on any accepted method of birth control or were lactating, uncontrolled hypertension (systolic >180 mm Hg, diastolic >115 mm Hg), significant cardiac, respiratory, renal, neurological, gastrointestinal, or endocrine disease, pre-existing cholelithiasis, pancreatic disease, history of gastrointestinal surgery for weight loss purposes, active gastrointestinal disease, drug treated diabetes mellitus, HbA₁c >10%, serum triglycerides 10 mmol/l, serum creatinine >150 mol/l, psychiatric disorders requiring medication that may impact upon the requirements of the protocol, history or presence of carcinoma, history or presence of bulimia or laxative abuse, clinical symptoms of fat soluble vitamin deficiencies, use/dependence of any substance of abuse, unable or unwilling to comply with the demands of the protocol, or subjects who had received any investigational drug within the previous 2 months). All participants gave their written informed consent. The study was conducted according to the Helsinki Declaration and the Ethical Committee in each of the four countries approved the protocol. The study was conducted between November 1998 and October 2002.

Study design

Before randomization the participants followed a VLED for 8 weeks with the aim of achieving a weight loss of 5%. The VLED provided 2.5–3.3 MJ (600–800 kcal) per day. One of two liquid protein preparations depending on commercial availability (Modifast; Novartis, Basel, Switzerland or Nutrilett; Nycomed Pharma, Oslo, Norway) was used in the study, provided free of charge. Of 383 participants, 309 achieved a 5% weight loss and were randomly assigned to 36 months of treatment with orlistat capsules (120 mg three times daily) or matching placebo capsules. Of these, 306 completed screening assessments of eating behavior and were eligible for this study, while three subjects refused to fill in the questionnaires.

After the VLED, participants were instructed to follow a diet reduced in energy by 2.5 MJ (600 kcal), mostly focused on reducing the fat content and following as closely as possible the prescription of 30% of total kcal from fat, 50% from carbohydrate, and 20% from protein. The energy deficit was individually determined and subtracted from an estimate of total energy expenditure based on the WHO formula (19) for calculating resting metabolic rate and was corrected for the physical activity level. Resting metabolic rate was estimated based on the subjects' weight after the VLED prior to randomization. According to the protocol a physical activity level of 1.3 or 1.5 was used for middle to moderate or heavy daily activity, respectively. A dietitian helped the participants to plan a diet at the appropriate energy level according to the abovementioned recommendations and the food preferences of the subjects. To standardize the dietary counseling, the 20 dietitians involved in the study were trained at two investigator meetings prior to the study. Based on individual preferences, the counseling was provided individually or in groups in sessions lasting 30 min. A dietitian who was blind to the drug assignment monitored patients every month for the first 18 months and then at 3-month intervals. The behavioral treatment was based on a structured program that included assignments to be completed at home (20). The subjects assessed problematic eating habits and were then given opportunities to suggest behavioral changes that were discussed with the dietitian or group participants. If the patients did not have suggestions for behavioral changes themselves, the dietitians suggested changes. The behavioral changes concerned food choices and the organization of meals as well as dealing with emotional states or specific situations and eating occasions.

Assessments

The body weight, waist circumference, and blood pressure were measured by standardized techniques (18). Fasting (10 hour overnight) blood samples were collected according to the protocol of the trial and laboratory parameters were analyzed centrally (Medilab, Copenhagen, Denmark).

Eating behavior was assessed with the TFEQ and the BES at screening (i.e., before VLED treatment) and at month 1, month 8, month 17, month 21, and month 33 after randomization. The TFEQ is a 51-item instrument, which contains three subscales measuring dietary restraint, disinhibition, and hunger. It consists of 36 closed questions with a forced, false/true answer format and 15 with a 4-point "Likert" scale (11). Dietary restraint is measured with questions like:

    How frequently do you avoid "stocking up" on tempting foods? Using a Likert scale, the participants are asked to respond to the most appropriate alternative of almost never, seldom, usually or almost always.

Disinhibition is measured using a false/true format including items such as:

    When I feel anxious, I find myself eating.

Hunger is measured using a false/true format, including items such as:

    I am usually so hungry that I eat more than three times a day.

The BES consists of 16 sets of 3–4 statements relating to binge eating behavior and the subject is asked to endorse the one item from each set which best describes the eating behavior. Scores were calculated and the ranges for possible scores were 0–21 for dietary restraint, 0–16 for disinhibition 0–14 for hunger and 0–46 for BES (12). Higher scores reflect a greater tendency to exhibit the particular eating behavior characteristics.

Statistical analyses

Descriptive statistics are presented as means s.d. and means (95% CI). Scores for eating behavior were calculated at screening time (i.e., before VLED treatment) and at months 1, 8, 17, 21, and month 33. Analyses of variance using least square means with changes in weight as covariates was done to test the possible difference between the orlistat and placebo group with regard to changes in scores for eating behavior at the measured time-points. Changes were calculated as the difference in scores between screening and subsequent measurements at months 1, 8, 17, 21, and 33. To test the interaction effect between treatment and gender, F -tests from ANCOVA were used. Student's t -test was used to test differences between gender and the Chi squared test was used in comparisons of categorical variables.

Pearson's correlations were done between changes in weight and changes in the scores for eating behavior within each treatment group at month 17 and month 33. To test the association between eating behavior and change in weight at month 33, we conducted multiple regression analyses with change in weight as the dependent variable and BMI at screening, age, gender, treatment, and the changes in scores for restraint, disinhibition, hunger, and binge eating, respectively, as four separate independent variables. Analyses were done using the observed values (for which the data are shown), last observation carried forward (from month 8) values and for the completers. The statistician (Fredrik Hansson) chose to use month 8 from screening (month 6 from randomization) because a reasonable test of the study medication required that it had to be taken for a relatively long period of time. The tests were considered significant at P < 0.05. Statistical analyses were performed using SAS software (version 9.1.3; SAS Institute, Cary, NC).

Power calculations was based on the main hypothesis of the study that the proportion of subjects with a maintained weight loss of at least 5% after the VLED was greater with orlistat than placebo treatment. The sample size calculation was performed using a formula from Lachin (21) N = (Z + Z )² 4 P (1 -P)/(P _c -P _e)² where Z is the significance level set to be 0.05, Z is the power of the study set to be 90%, P _c is the proportion of treatment success in the orlistat group set to be 0.60, P _e is the proportion of treatment success in the placebo group set to be 0.35 and P is the overall treatment success rate. With this, the total number of subjects needed would be 168. With a drop out rate of 40% the number of subjects required would be 280. The statistical power of the study to achieve this goal was 92%.

Results

Of the 306 subjects whose eating behavior was assessed at screening, 251 (82.0%) and 197 (64.4%) subjects remained in the trial and completed eating behavior assessments at months 17 and 33, respectively. Within each group, 1 subject was unwilling to complete the BES following screening. The last observation carried forward population consisted of 140 subjects in the orlistat group and 137 subjects in the placebo group and the completer population included 103 subjects and 98 subjects in the orlistat and placebo group, respectively. Table 1 shows the subject characteristics. These were similar in the orlistat and placebo groups.

Table 1 - Characteristics of the participants in the orlistat and placebo groups before weight reduction (screening).

Full table

Eating behavior

Table 2 shows the eating behavior scores at screening and the changes in eating behaviors at month 1, 8, 17, 21, and 33 in the orlistat and placebo groups. No difference was seen between the groups before weight reduction. The mean scores from the BES tended to be low.

Table 2 - Scores before weight reduction (screening) and changes in eating behavior in the orlistat and placebo group; the observed population.

Full table (113K)

The main changes in eating behavior occurred between screening and month 1 and remained quite stable thereafter (Table 2). There was no significant difference between genders with regard to change in dietary restraint, disinhibition and binge eating (data not shown). Dietary restraint increased, while disinhibition, hunger, and binge eating decreased with no between-group differences with the exception of the score for hunger. The scores for hunger were reduced more in the placebo than in the orlistat group at month 33 and this was statistically significant in men (between-group difference -1.1 (95% CI -1.2, 0) P = 0.0493). This tendency was similar in women in numerical terms but did not achieve statistical significance (between-group difference -1.0 (95% CI -2.2, 0.3) P = 0.1395). There was no statistically significant interaction effect between treatment and gender (Table 2). However, scores for restraint increased more in men than in women (7.0 (95% CI 6.3, 7.8) vs. 5.0 (95% CI 4.3, 5.7), respectively; P = 0.0002) but there was no difference between the genders at month 33 (5.8 (95% CI 4.9, 6.6) vs. 5.0 (95% CI 4.2, 5.8), P = 0.2).

Change in weight was significantly associated with change in restraint, disinhibition and binge eating at month 17 and 33 and with hunger at month 17 (all P < 0.01), but not with hunger at month 33 (P = 0.09). The same trends were seen in the last observation carried forward analyses and in the analyses of completers (data not shown).

Table 3 shows the correlations between change in weight and eating behavior within each group in the observed population. In the orlistat group, change in weight was inversely correlated to change in restraint and positively correlated to change in disinhibition and binge eating. In the placebo group, change in weight was inversely correlated to change in restraint. In addition change in hunger and binge eating was positively correlated to change in weight at month 17. The correlations observed in the completer analyses as well as the last observation carried forward analyses were very similar.

Table 3 - Correlations between change in weight and eating behavior in the orlistat and placebo group (observed population).

Full table (27K)

The results of the multiple linear regression analyses for restraint, disinhibition, hunger and binge eating, each adjusted for treatment, BMI at screening, gender, and age are shown in Table 4.Changes in restraint, disinhibition, and binge eating were significantly associated with weight loss (all P 0.002), explaining 12–17% of the variation in weight predicted for month 33. Changes in hunger tended to be associated with weight loss (P = 0.0534) explaining 9% of the variation. We also included smoking as an independent variable, but smoking was not a significant predictor in the models (data not shown).

Table 4 - Multiple regression analyses of change in eating behavior scores as predictors for weight change at month 33 in 197 subjects in the observed population.

Full table

Discussion

In this 3-year, placebo-controlled clinical trial of weight maintenance following a VLED we found no substantial differences between the effect of orlistat or placebo on eating behavior measured with the TFEQ and BES. Increased restraint and decreased disinhibition and binge eating were predictors of sustained weight maintenance in these participants; the participants were obese men and women with metabolic syndrome risk factors. Thus, changes in eating behavior in the orlistat group are mostly due to quantified eating behavior similar to the way the placebo operates. We are not aware of previous reports that have examined the effect of long-term orlistat treatment on eating behavior following a VLED.

Compared to scores at screening, dietary restraint was increased and disinhibition, hunger and binge eating were decreased in both the orlistat and placebo groups. An improvement in restraint, disinhibition and hunger has been shown in other trials of obese subjects taking sibutramine, an obesity drug that influences appetite (22,23). In the study by Hainer et al. 80 women were followed for 1 year and a drop in the disinhibition score was the most important factor associated with the decrease in BMI (22). In the study by Bauer et al., the improvement in eating behavior was mostly attributed to the cognitive weight loss program (23). This may have been the case in our study, too, as all the subjects took part in a dietitian-managed behavioral program.

We found no differences between the treatment with orlistat or with the placebo with regard to restraint, disinhibition and binge eating. Thus, this form of drug treatment appears to be safe with regard to eating behavior amongst obese adults. Notably the placebo group showed a greater reduction in hunger scores than the orlistat group at 33 months. This difference may be due to chance or to the greater degree of weight loss in orlistat-treated subjects though a 2 kg difference between the groups seems unlikely to affect regulators of appetite. Energy balance and appetite regulation are controlled by a number of hormones and peptides (24). Decreased levels of leptin may increase the release of orexigenic signals and thereby induce hunger (25). Unfortunately, we did not measure leptin levels in this study. The difference in hunger between the orlistat and the placebo group may be of minor clinical relevance given the overall reduction in hunger.

Increase in dietary restraint was inversely correlated to weight maintenance in both groups. However, the associations between change in weight and other eating behaviors differed somewhat within the orlistat and placebo groups. Lower scores for disinhibition and binge eating were associated with changes in weight only in the orlistat group. High scores for disinhibition and hunger have been associated with a high intake of fatty foods (26) and with a greater binge eating severity (14,27,28). During binge eating the energy intake per episode may be of several thousand kJ and highly palatable, fatty foods are preferred (29). Participants taking orlistat may have experienced gastric side effects when eating palatable and fatty food, consistent with its mechanism of action. Though the overall percentage of energy from fat did not differ between the groups (18,30), a subgroup of participants may have reduced the intake of energy yielding fatty foods and improved their weight maintenance because of that. Indeed, better weight loss and reduced binge eating has been reported in obese binge eaters treated with orlistat (31). Other contributors to weight loss maintenance than those measured here could be significant predictors of success. Despite the blinded design of the study, some participants may have guessed that they received the placebo. This may have increased their motivation to reach weight goals or maintain a certain amount of weight loss (32).

We further sought to understand the role of eating behavior in long-term weight maintenance. After adjustment, increase in restraint and decrease in disinhibition and binge eating were all associated with improved weight maintenance, as expected, given that any dietary treatment of obesity implies increased restraint. Obese individuals who are not in treatment tend to score similarly to non-obese individuals as regards cognitive restraint; however, their scores increase during treatment (33). Likewise, obese subjects score higher than non-obese subjects with regard to disinhibition but their scores fall during treatment (13). Moreover, scores for disinhibition have been shown to correlate with the severity of binge eating (34). The relative importance of restraint, disinhibition, hunger, and binge eating as predictors for weight maintenance has been shown in a number of other studies (16,17,27,35,36). In the study of Pekkarinen et al. obese subjects were followed for 2 years after a weight loss program of 17 sessions including an 8-week VLED (16). Subjects that maintained an increase in restraint and decreases in disinhibition and hunger also were most successful with their weight reduction (16). Among subjects who were followed for 1 year after a VLED-induced initial weight loss, success was associated with increased dietary restraint and lesser feelings of hunger (17). In a study by Westenhoefer et al. 7407 men and women participated in a computer-based weight loss program. Successful weight reduction was associated with higher baseline scores for restraint that increased further during treatment and a lower baseline score of disinhibition that was further reduced (36). Changes in BES were not measured in these studies.

During weight maintenance it may be of special importance to focus on increasing dietary restraint and reducing disinhibition. This is challenging in an environment where there is an abundance of fatty foods and snacks, but inter-personal group approaches for dealing with stimuli control and disinhibitors such as lack of time, social eating, stress, anxiety, distress, and loneliness may be helpful (11).

Our study has several strengths. With over three hundred obese individuals and about equal numbers of men and women, its statistical power was good. The subjects were followed for 3 years in a placebo-controlled trial and the completion rate was relatively high. We were able to prospectively measure changes in eating behavior at several time-points. One of the limitations is that we did not perform the assessments of eating behavior at randomization, after the end of the VLED. It was not considered to be clinically relevant to measure eating behavior in subjects at the end of a VLED fast before a normal diet was restarted. Thus, we were unable to assess changes during the VLED. However, the changes in restraint, disinhibition and hunger were in accordance with others (16,17,35). Moreover, no difference in weight loss was seen between the orlistat and placebo groups after VLED (18).

In conclusion, our novel finding was that, in a long-term weight maintenance trial, orlistat did not affect eating behavior in a substantially different way than did the placebo. These findings indicate that orlistat does not impair the effect of a behavioral program and that changes in eating behavior seem to be important determinants of success during drug treatment.

Disclosure

Serena Tonstad has received honoraria for lectures from Hoffman La Roche, the manufacturer of orlistat.

References

REFERENCES

1. Laaksonen DE, Lakka HM, Niskanen LK et al. Metabolic syndrome and development of diabetes mellitus: application and validation of recently suggested definitions of the metabolic syndrome in a prospective cohort study. Am J Epidemiol 2002;156:1070–1077.  | Article | PubMed | ISI |
2. Lakka HM, Laaksonen DE, Lakka TA et al. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 2002;288:2709–2716.  | Article | PubMed | ISI |
3. Lemieux I, Pascot A, Couillard C et al. Hypertriglyceridemic waist: a marker of the atherogenic metabolic triad (hyperinsulinemia; hyperapolipoprotein B; small, dense LDL) in men? Circulation 2000;102:179–184.  | PubMed | ISI | ChemPort |
4. Tanko LB, Bruun JM, Alexandersen P et al. Novel associations between bioavailable estradiol and adipokines in elderly women with different phenotypes of obesity: implications for atherogenesis. Circulation 2004; 110:2246–2252.  | Article | PubMed | ChemPort |
5. Ayyad C, Andersen T. Long-term efficacy of dietary treatment of obesity: a systematic review of studies published between 1931 and 1999. Obes Rev 2000;1:113–119.  | Article | PubMed | ChemPort |
6. Wadden TA, Butryn ML, Byrne KJ. Efficacy of lifestyle modification for long-term weight control. Obes Res 2004; 12(Suppl):151S–162S.
7. Sjöström L, Rissanen A, Andersen T et al. Randomised placebo-controlled trial of orlistat for weight loss and prevention of weight regain in obese patients. European Multicentre Orlistat Study Group. Lancet 1998;352:167–172.  | Article | PubMed | ISI | ChemPort |
8. Hill JO, Hauptman J, Anderson JW et al. Orlistat, a lipase inhibitor, for weight maintenance after conventional dieting: a 1-y study. Am J Clin Nutr 1999; 69:1108–1116.  | PubMed | ISI | ChemPort |
9. Pirke KM, Laessle RG. Restrained eating. In: Stunkard AJ, Wadden TA. (eds). Obesity: Theory and Therapy, 2nd edn.Raven Press: New York, 1993, pp 151–162.
10. Lowe MR, Maycock B. Restraint, disinhibition, hunger and negative affect eating. Addict Behav 1988;105:145–150.
11. Stunkard AJ, Messicks S. The Three Factor Eating Questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res 1985;29:71–83.  | Article | PubMed | ISI | ChemPort |
12. Gormally J, Black S, Daston S, Bardin D. The assessment of binge eating severity among obese persons. Addict behav 1982; 7:47–55.  | Article | PubMed | ISI | ChemPort |
13. Clark M, Marcus V, Pera V, Niaura R. Changes in Eating Inventory scores following obesity treatment. Int J Eat Dis 1994; 15:401–405.  | ChemPort |
14. 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 1998;22:778–785.  | Article | ChemPort |
15. Cooper Z, Hawker DM, Fairrburn CG. Cognitive-behavioral Treatment of Obesity: A Clinician's Guide.The Guilford Press: New York, 2003.
16. Pekkarinen T, Takala I, Mustajoki P. Two year maintenance of weight loss after a VLCD and behavioural therapy for obesity: correlation to the scores of questionnaires measuring eating behaviour. Int J Obes 1996;20:332–337.  | ChemPort |
17. Vogels N, Westerterp-Plantenga MS. Categorical strategies based on subject characteristics of dietary restraint and physical activity, for weight maintenance. Int J Obes 2005;29:849–857.  | Article | ChemPort |
18. Richelsen B, Tonstad S, Rössner S et al. Effect of orlistat on weight regain and on cardiovascular risk factors following a very-low-calorie diet in abdominally obese patients. A three-year-randomized placebo controlled study. Diabetes Care 2007; 30:27–32.  | Article | PubMed | ChemPort |
19. FAO/WHO/UNU. Energy and Protein Requirements. Geneva: World Health Organization, 1985.
20. Melin I, Rössner S. Practical clinical behavioral treatment of obesity. Patient Educ Counc 2003;49:75–83.  | Article |
21. Lachin JM. Introduction to sample size determination and power analysis for clinical trials. Control Clin Trials 1981;93:93–113.  | Article |
22. Hainer V, Kunesova M, Bellisle F et al. Psychobehavioral and nutritional predictors of weight loss in obese women treated with sibutramine. Int J Obes 2005; 29:208–216.  | Article | ISI | ChemPort |
23. Bauer C, Fischer A, Keller U. Effect of sibutramine and cognitive-behavioral weight loss therapy in obesity and subclinical eating disorders. Diab Obes Metab 2006;8:289–295.  | Article | ChemPort |
24. Huda MSB, Wilding JPH, Pinkney JH. Appetite regulatory peptides. Gut peptides and the regulation of appetite. Obes Rev 2006;7:163–182.  | Article | PubMed | ChemPort |
25. Blundell JE. Perspective on the central control of appetite. Obesity 2006; 14(Suppl 4):160S–163S.
26. Lindroos AK, Lissner L, Mathiassen ME et al. Dietary intake in relation to restrained eating, disinhibition, and hunger in obese and nonobese Swedish women. Obes Res 1997;5:175–182.  | PubMed | ISI | ChemPort |
27. Marcus MD, Wing RR, Lamparski DM. Binge eating and dietary restraint in obese patients. Addict behav 1985; 10:163–168.  | Article | PubMed | ChemPort |
28. Wadden TA, Foster GD, Letizia KA, Wilk JE. Metabolic, anthropometric and psychological characteristics of binge eaters. Int J Eat Disorder 1993;14:17–25.  | Article | ChemPort |
29. Marcus MD. Binge eating in obesity. In: Fairburn CG, Wilson GT. (eds). Binge Eating: Nature, Assessment and Treatment.Guilford Press: New York, 1993, pp 77–96.
30. Svendsen M, Helgeland M, Tonstad S. Does long-term treatment with orlistat influence dietary fat intake in obese subjects with metabolic syndrome risk factors? (Abstract) Int J Obes Relat Metab Disord 2003; 27(Suppl 1):S123
31. Golay A, Laurent-Jaccard A, Habicht F et al. Effect of orlistat in obese patients with binge eating disorder. Obes Res 2005; 13:1701–1708.  | PubMed | ChemPort |
32. Foster GD, Wadden TA, Voght RA, Brewer G. What is a reasonable weight loss? Patients' expectations and evaluations of obesity treatment outcomes? J Consult Clin Psychol 1997;65:79–85.  | Article | PubMed | ISI | ChemPort |
33. Björvell H, Rössner S, Stunkard AJ. Obesity, weight loss and dietary restraint. Int J Eat Dis 1986;5:727–734.  | Article |
34. Lawson OJ, Williamson DA, Champagne CM et al. The associations of body weight, dietary intake and energy expenditure with dietary restraint and disinhibition. Obes Res 1995; 3:153–161.  | PubMed | ISI | ChemPort |
35. Lejeune MPGM, Hukshorn CJ, Saris WHMWesterterp-Plantenga MS. Effect of dietary restraint during and following pegylated recombinant leptin (PEG-OB) treatment of overweight men. Int J Obes 2003;27:1494–1499.  | Article | ChemPort |
36. Westenhoefer J, Stunkard AJ, Pudel V. Validation of the flexible and rigid control dimensions of dietary restraint. Int J Eat Dis 1999;26:53–64.  | Article | ChemPort |

Acknowledgments

We thank the subjects for participating and the staff who were involved in this study. Hoffman La Roche financed the trial. Serena Tonstad has received honoraria for lectures from Hoffman La Roche, the manufacturer of orlistat.