SEARCH

SEARCH BY CITATION

Keywords:

  • gender differences;
  • Three-Factor Eating Questionnaire;
  • weight management;
  • body fat distribution;
  • food intake

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Objective: To put into relationship the dietary and anthropometric profile of men and women with their eating behaviors (cognitive dietary restraint, disinhibition, and susceptibility to hunger) and to assess whether gender and obesity status influence these associations.

Research Methods and Procedures: Anthropometric measurements (including visceral adipose tissue accumulation), dietary profile (3-day food record), and eating behaviors (Three-Factor Eating Questionnaire) were determined in a sample of 244 men and 352 women.

Results: Women had significantly higher cognitive dietary restraint and disinhibition scores than men (p < 0.0001). In both genders, scores for disinhibition and susceptibility to hunger, but not for cognitive dietary restraint, were higher in obese subjects than in overweight and nonobese subjects (p < 0.05). Positive correlations were observed between rigid restraint and most of the anthropometric variables studied (0.12 ≤ r ≤ 0.16). Moreover, in women, flexible restraint was negatively associated with body fat and waist circumference (r = −0.11). Cognitive dietary restraint and rigid restraint were positively related to BMI among nonobese women (0.19 ≤ r ≤ 0.20), whereas in obese men, cognitive dietary restraint and flexible restraint tended to be negatively correlated with BMI (−0.20 ≤ r ≤ −0.22; p = 0.10).

Discussion: Gender could mediate associations observed between eating behaviors and anthropometric profile. It was also found that disinhibition and susceptibility to hunger are positively associated with the level of obesity. On the other hand, cognitive dietary restraint is not consistently related to body weight and adiposity, whereas rigid and flexible restraint are oppositely associated to obesity status, which suggests that it is important to differentiate the subscales of cognitive dietary restraint. Finally, counseling aimed at coping with disinhibition and susceptibility to hunger could be of benefit for the long-term treatment of obesity.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Obesity is a major public health problem in developed countries, and its prevalence has steadily risen (1, 2, 3). Previous studies have shown that nutritional education combined with eating behavior therapy is necessary to improve the effectiveness of obesity treatments (4). To identify eating behaviors related to obesity and the management of body weight, questionnaires have been developed to describe human eating behaviors. Over the last decade, one of the most widely used scales in behavioral research has been the Three-Factor Eating Questionnaire (TFEQ)1 that was constructed by Stunkard and Messick in 1985 (5). This questionnaire was developed to measure cognitive dietary restraint, disinhibition, and susceptibility to hunger. More recently, other researchers have used the TFEQ to define more specific subscales of eating behaviors (6, 7).

Numerous studies have examined eating behaviors in relation to gender, obesity status, dietary patterns, and weight loss success (8). In general, studies have shown that women have higher cognitive dietary restraint and disinhibition scores than men (9, 10). Studies have also reported that obese subjects generally display higher scores for disinhibition (6, 9, 11, 12, 13) and susceptibility to hunger (13, 14, 15) than nonobese individuals. It also has been suggested that disinhibition and susceptibility to hunger are positively associated with energy intake (11, 13, 16), whereas cognitive dietary restraint is generally negatively correlated with energy and dietary fat intakes (13, 17, 18). Studies have reported that higher cognitive dietary restraint seems to be a predictor of weight loss (12, 14, 19), whereas disinhibition would be the major determinant of weight gain or weight regain after weight loss (12, 14). However, most of these studies have been conducted in individuals who were enrolled in controlled weight loss interventions and not in a “real-life” uncontrolled context. Furthermore, to our knowledge, the modulating effects of gender and obesity status on the associations of eating behaviors with anthropometric profile and food patterns have not been well documented. Finally, even if numbers of papers have reported data on eating behaviors, only few studies so far have examined subscales of eating behaviors as described by Westenhoefer et al. (6) and Bond et al. (7). These subscales are of particular interest because they allow refinement of definitions of eating behaviors by representing distinct aspects of cognitive dietary restraint, disinhibition, and susceptibility to hunger.

The main purpose of this cross-sectional study was to address whether eating behaviors and their subscales would be associated with anthropometric variables and dietary intakes in a sample of men and women involved in the Québec Family Study (QFS). More specifically, it was hypothesized that, in subjects not involved in a controlled weight loss intervention, cognitive dietary restraint may not be related to the anthropometric profile, whereas its subscales (flexible and rigid restraint) could be oppositely associated with adiposity and fat distribution indexes. It was also hypothesized that gender and obesity status could modulate the associations of eating behaviors and their subscales with anthropometric profile and dietary intakes.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Subjects

The QFS was initiated at Laval University in 1978 (20). The primary objective of this study was to investigate the role of genetics in the etiology of obesity and related cardiovascular risk factors. The families involved were recruited through the media and were all French Canadians. The participation was voluntary, and all subjects signed an informed-consent document. The study was approved by the Medical Ethics Committee of Laval University.

TFEQ

A French version of the TFEQ was filled out by 596 healthy adults (244 men and 352 women) from QFS. The TFEQ is a 51-item questionnaire developed by Stunkard and Messick in 1985 (5). Its purpose is to assess three factors related to cognitions and behaviors associated with eating. These factors are cognitive dietary restraint, disinhibition, and susceptibility to hunger. More precisely, cognitive dietary restraint is the intent to restrict food intake to control body weight (21 items; score ranging from 0 to 21). Disinhibition is an overconsumption of food in response to a variety of stimuli, such as emotional stress, associated with a loss of control on food intake (16 items, score ranging from 0 to 16). Finally, susceptibility to hunger refers to food intake in response to feelings and perceptions of hunger (14 items; score ranging from 0 to 14) (5). This questionnaire has been validated, and all three of its scales had good test-retest reliability (5, 21, 22).

More specific subscales for these three general eating behaviors have also been determined, as suggested by Westenhoefer et al. (6) and Bond et al. (7). First, cognitive dietary restraint has been divided into rigid and flexible restraint (7 items for each subscale; score ranging from 0 to 7), as proposed by Westenhoefer and colleagues (6). Rigid restraint is defined as a dichotomous, all-or-nothing approach to eating, dieting, and weight, whereas flexible restraint would be a more gradual approach to eating, dieting, and weight in which, for example, “fattening” foods are eaten in limited quantities without feelings of guilt (6).

Disinhibition has also been divided into three subscales (habitual, emotional, and situational susceptibility to disinhibition) as suggested by Bond and colleagues (7). Habitual susceptibility to disinhibition describes behaviors that may occur when circumstances could predispose to recurrent disinhibition (5 items; score ranging from 0 to 5). Emotional susceptibility to disinhibition defines a kind of disinhibition that is associated with negative affective states (3 items; score ranging from 0 to 3), whereas situational susceptibility to disinhibition refers to a type of disinhibition that is initiated by specific environmental cues (5 items; score ranging from 0 to 5). Susceptibility to hunger has also been divided into two specific subscales (internal and external locus for hunger) (6 items for each subscale; score ranging from 0 to 6) as proposed by Bond and coworkers (7). Internal hunger refers to the type of hunger that is interpreted and regulated internally, whereas external hunger is triggered by external cues (7).

Dietary Profile

Subjects quantified foods and drinks consumed using a 3-day estimated food record, which included 2 week days and 1 weekend day. A nutritionist explained to each subject how to complete their 3-day food record and encouraged them to continue to consume usual amounts of typical foods and beverages. After completion of the record, the nutritionist reviewed it with the participant, and nutrient intakes were calculated with a computerized version of the Canadian Nutrient File (23). To take into account the possibility of underreporting, cutoff limits were used, as proposed by Goldberg et al. (24). These cutoff limits have been developed to identify subjects with a reported energy intake below 1.35× basal metabolic rate (BMR) because such an intake cannot be representative of long-term habitual intakes. A measurement of resting metabolic rate (RMR) was performed by indirect calorimetry for all subjects involved in our study. Although an RMR value for a given subject is generally slightly higher than BMR value, a recent study has shown that the Schofield equations overestimated BMR value; thus, this predicted BMR was unsuitable, especially among obese populations (25). Accordingly, in our sample, predicted BMR was slightly higher than measured RMR (data not shown). For all these reasons, it was decided to use RMR in our calculations because this value was actually measured, whereas BMR would have been estimated. Therefore, in our sample, subjects who had reported energy intakes that fell below 1.35× RMR were excluded from the analyses involving dietary variables. The percentage of energy (kilojoules) derived from proteins, carbohydrates, and dietary fat was also calculated.

Indexes of Adiposity and of Body Fat Distribution

Body weight, height, and waist circumference were measured according to standardized procedures recommended at the Airlie Conference (26), and BMI was calculated. RMR was obtained by indirect calorimetry measurement. Measurements were performed over a 30-minute period by using an open-circuit ventilated-hood system. Body density was obtained from the mean of six measurements from the hydrostatic weighing technique (27). Before immersion in the hydrostatic tank, the helium dilution method of Meneely and Kaltreider (28) was used to determine the pulmonary residual volume. The percentage of total body fat was determined from body density with the equation of Siri (29). Finally, to quantify abdominal fat accumulation and cross-sectional areas of visceral and subcutaneous accumulation, abdominal adipose tissue areas were assessed by computed tomography using a Siemens Somatom DRH scanner (Elanger, Germany) as described previously (30). The computed tomography scan was performed between L4 and L5 vertebrae.

Statistical Analysis

Student's t test was used to compare means of physical characteristics, dietary profile, and eating behaviors between men and women. To determine the obesity status of our subjects, we classified subjects into three categories of BMI. Subjects with a BMI below 25 were considered as being nonobese. Those with a BMI between 25 and 30 were classified as overweight, and subjects with a BMI ≥ 30 kg/m2 were considered obese. Differences regarding the distribution of the obesity status among men (32% nonobese, 41% overweight, and 27% obese) and women (41% nonobese, 25% overweight, and 34% obese) were observed (χ2 = 17.1; p = 0.0002), which justified the performance of further analyses according to the obesity status. Therefore, an ANOVA was performed to assess differences between means of eating behaviors in nonobese, overweight, and obese men and women. In the presence of a significant group effect, Duncan's tests were performed to determine which of the groups were significantly different. To quantify the univariate relationships of eating behaviors and their subscales with anthropometric variables, Spearman correlations were performed in both men and women and in subgroups formed on the basis of the obesity status. Because significant associations between eating behaviors and BMI have been reported consistently in the literature (6, 9, 11, 13), it was decided, therefore, to perform BMI-adjusted correlations between eating behaviors and dietary variables. The probability level for significance used for the interpretation of all statistical analyses was set at an alpha level of p < 0.05. All analyses were performed by using SAS statistical software (SAS Institute, Cary, NC).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Table 1 shows that men had higher weight, waist circumference, and higher visceral adipose tissue (VAT) than women, whereas women had greater body fat and subcutaneous adipose tissue (SAT) than men. Moreover, in subjects who reported valuable energy intakes, it can be observed that men had higher energy intake than women.

Table 1.  Physical characteristics and dietary profile of men and women of the study
 Men (n = 244)Women (n = 352)
  • Values are means ± SD. For body fat mass, n = 202 men and 289 women; for visceral and subcutaneous adipose tissue, n = 197 men and 292 women; dietary variables are reported only for subjects with valuable energy intakes, n = 176 men and 225 women.

  • *

    Significantly different from men, p < 0.0001.

  • Significantly different from men, p < 0.01.

  • Percentage of proteins, dietary fat, and carbohydrates = percentage of energy (kilojoules) derived from these macronutrients.

Age (years)43.4 ± 14.942.0 ± 14.6
Weight (kg)85.2 ± 22.073.4 ± 20.8*
BMI (kg/m2)28.5 ± 7.028.8 ± 8.2
Body fat (%)23.9 ± 9.433.0 ± 10.2*
Waist girth (cm)96.4 ± 17.786.8 ± 18.6*
VAT (cm2)138 ± 89110 ± 76
SAT (cm2)224 ± 135354 ± 185*
Energy (MJ)12.5 ± 2.99.4 ± 2.0*
Proteins (%)15.6 ± 3.016.0 ± 3.0
Dietary fat (%)34.4 ± 5.934.7 ± 6.1
Carbohydrates (%)47.5 ± 6.947.4 ± 6.7

Results presented in Table 2 indicate that women had higher cognitive dietary restraint and disinhibition scores and higher scores for flexible and rigid restraint and higher scores for habitual and emotional susceptibility to disinhibition than men. The predominance of rigid restraint behaviors, which was determined by dividing rigid restraint score by the summation of rigid and flexible restraint scores, was also higher in women than in men.

Table 2.  Differences in cognitive dietary restraint, disinhibition, and susceptibility to hunger between men and women
 Men (n = 244)Women (n = 352)
  • *

    Significantly different from men, p < 0.0001.

  • Rigid/total restraint = rigid restraint divided by the summation of rigid and flexible restraint.

  • Values are means ± SD. Restraint: intent to control food intake; rigid restraint: dichotomous approach to eating; flexible restraint: gradual approach to eating; disinhibition: overconsumption of food in response to cognitive or emotional cues; habitual susceptibility: recurrent disinhibition; emotional susceptibility: negative affective states; situational susceptibility: environmental cues; susceptibility to hunger: food intake in response to feelings and perceptions of hunger; internal hunger: interpreted and regulated internally; and external hunger: triggered by external cues.

Cognitive dietary restraint5.8 ± 3.58.4 ± 4.7*
 Flexible restraint2.2 ± 1.43.0 ± 1.8*
 Rigid restraint1.2 ± 1.42.3 ± 1.8*
 Rigid/total restraint0.3 ± 0.30.4 ± 0.2*
Disinhibition4.6 ± 3.05.8 ± 3.3*
 Habitual susceptibility0.6 ± 1.01.1 ± 1.3*
 Emotional susceptibility0.6 ± 1.01.4 ± 1.3*
 Situational susceptibility1.9 ± 1.51.9 ± 1.5
Susceptibility to hunger4.3 ± 3.53.9 ± 3.1
 Internal hunger1.6 ± 1.91.4 ± 1.7
 External hunger1.8 ± 1.51.7 ± 1.5

It can be seen in Table 3 that disinhibition in both genders and susceptibility to hunger in women were positively associated with all anthropometric variables studied. Susceptibility to hunger in men was significantly associated with BMI, percentage of body fat, and waist circumference. Cognitive dietary restraint was not associated with any of the anthropometric variables studied in either men or women. However, when cognitive dietary restraint was divided into flexible and rigid restraint, a positive correlation was found in men between rigid restraint and SAT area. In women, rigid restraint was positively and significantly related to all anthropometric variables studied. Moreover, flexible restraint was negatively correlated with body fat and waist circumference, but only in women.

Table 3.  Spearman's correlation coefficients for the associations between eating behaviors and adiposity/fat distribution indexes in men and women
  • For body fat mass, n = 202 men and n = 289 women; for VAT and SAT, n = 197 men and n = 292 women. Refer to Table 2 for a detailed description of eating behaviors presented in this table.

  • *

    Significant correlation, p < 0.05.

  • Significant correlation, p < 0.0001.

  • Significant correlation, p < 0.01.

 Men (n = 244)
 BMIBody fatWaist girthVATSAT
Restraint0.030.01−0.010.010.06
 Flexible restraint−0.06−0.10−0.09−0.07−0.06
 Rigid restraint0.110.110.090.120.16*
Disinhibition0.490.380.460.250.42
Hunger0.280.16*0.250.090.13
 Women (n = 352)
 BMIBody fatWaist girthVATSAT
Restraint0.050.040.010.010.02
 Flexible restraint−0.05−0.11*−0.11*−0.08−0.07
 Rigid restraint0.150.12*0.12*0.12*0.12*
Disinhibition0.480.430.430.260.49
Hunger0.250.210.220.12*0.24

After exclusion of subjects who reported implausibly low energy intakes, actual energy intake expressed as a multiple of RMR (EI/RMR) was compared according to gender and BMI classes. No gender differences in EI/RMR were observed, but obese subjects had lower EI/RMR than nonobese subjects (obese = 1.7 vs. nonobese = 1.9; p < 0.05). Table 4 summarizes intercorrelations between eating behaviors and dietary variables after controlling for the potential confounding effect of BMI. Cognitive dietary restraint was negatively associated with reported energy intake and proportion of dietary fat in both genders. Only in women, flexible restraint was negatively associated with reported energy intake, whereas disinhibition and habitual, emotional, and situational susceptibility to disinhibition were positively related to reported energy intake and to the proportion of dietary fat in the diet. Susceptibility to hunger and internal hunger were positively associated with reported energy intake in both genders. Finally, external hunger was positively associated with the proportion of dietary fat in women, whereas internal hunger was positively related to proportion of dietary fat in men.

Table 4.  BMI-adjusted Spearman's correlation coefficients for the associations between eating behaviors and dietary variables in men (n = 176) and women (n = 225)*
 Energy (kcal)Dietary fat (%)
 WomenMenWomenMen
  • *

    Results are reported only for subjects who reported valuable energy intakes.

  • Dietary fat = percentage of energy derived from fat (%).

  • Significant correlation, p < 0.01.

  • §

    Significant correlation, p < 0.05.

  • Refer to Table 2 for a detailed description of eating behaviors presented in this table.

Restraint−0.17−0.19−0.15§−0.19§
 Flexible restraint−0.21−0.04−0.12−0.08
 Rigid restraint−0.04−0.13−0.10−0.08
Disinhibition0.230.080.220.09
 Habitual susceptibility0.16§0.030.220.13
 Emotional susceptibility0.230.030.220.07
 Situational susceptibility0.190.060.13§0.03
Hunger0.230.200.100.14
 Internal hunger0.200.18§0.020.15§
 External hunger0.120.100.170.10

Figure 1 shows that cognitive dietary restraint was not different among the three categories of BMI in both genders. However, obese men exhibited a higher disinhibition score than overweight and nonobese men. Obese women also displayed a higher disinhibition score than overweight women, who, in turn, had a higher score than nonobese women. Finally, both obese men and women had a higher degree of susceptibility to hunger than overweight and nonobese subjects.

image

Figure 1. Differences in cognitive dietary restraint, disinhibition, and susceptibility to hunger between nonobese, overweight, and obese men and women (n = 596). Bars within each eating behavior with different superscript letters are significantly different (p < 0.05). Values are means ± SE. Refer to Table 2 for a detailed description of eating behaviors presented in this figure.

Download figure to PowerPoint

Furthermore, it is shown in Figure 1 that all groups of women, irrespective of their obesity status, had higher scores of cognitive dietary restraint than nonobese, overweight, and obese men. Gender differences in disinhibition were also observed in nonobese and overweight subjects, with women having higher scores than men. Accordingly, a significant interaction between sex and BMI categories was observed for disinhibition (F = 5.84; p = 0.003). Regarding the susceptibility to hunger, obese men had significantly higher scores than obese women.

Figure 2 presents correlation coefficients for the associations between eating behaviors and BMI in nonobese, overweight, and obese men. Cognitive dietary restraint (r = −0.20) and flexible restraint (r = −0.22) tended to be negatively associated with BMI, but only in obese men. Disinhibition was positively related to BMI in nonobese (r = 0.29) and obese (r = 0.28) men. There was also a positive association between susceptibility to hunger and BMI in overweight (r = 0.20) and obese (r = 0.36) men, whereas no association was observed among nonobese men. In women (Figure 3), cognitive dietary restraint (r = 0.20) and rigid restraint (r = 0.19) were significantly associated with BMI, but only in nonobese women. Disinhibition was positively related to BMI among nonobese (r = 0.40), overweight (r = 0.26), and obese (r = 0.19) women. As in men, there was also a positive association between susceptibility to hunger and BMI, but only in overweight (r = 0.33) and obese (r = 0.21) women.

image

Figure 2. Spearman's correlation coefficients for the associations between eating behaviors and BMI in nonobese (n = 78), overweight (n = 100), and obese men (n = 66). Trend for a correlation: T, p < 0.10; significant correlation: p < 0.05 (*). Refer to Table 2 for a detailed description of eating behaviors presented in this figure.

Download figure to PowerPoint

image

Figure 3. Spearman's correlation coefficients for the associations between eating behaviors and BMI in nonobese (n = 146), overweight (n = 88), and obese women (n = 118). Significant correlation: p < 0.05 (*). Refer to Table 2 for a detailed description of eating behaviors presented in this figure.

Download figure to PowerPoint

Table 5 summarizes the intercorrelations among eating behaviors in men and women. In both genders, positive associations were observed for cognitive dietary restraint with flexible restraint and rigid restraint and for disinhibition with rigid restraint and susceptibility to hunger. Negative relationships were observed between cognitive dietary restraint and susceptibility to hunger, between disinhibition and flexible restraint, and between susceptibility to hunger and flexible restraint in men, whereas no such association was observed in women. Finally, cognitive dietary restraint was positively related to disinhibition, but only in women.

Table 5.  Intercorrelations among eating behaviors in men (n = 244) and women (n = 352)
 Flexible restraintRigid restraintDisinhibitionHunger
 MenWomenMenWomenMenWomenMenWomen
  • Refer to Table 2 for a detailed description of eating behaviors presented in this table.

  • *

    Significant correlation, p < 0.0001.

  • Significant correlation, p < 0.01.

  • Significant correlation, p < 0.05.

Restraint0.79*0.82*0.70*0.81*−0.040.13−0.210.01
Disinhibition−0.15−0.010.190.25*0.52*0.58*
Hunger−0.23−0.040.000.10

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Our finding that disinhibition and susceptibility to hunger were positively correlated with BMI and total body fat are concordant with other studies (6, 9, 11, 12, 13, 14, 15, 31). We found no relationship between cognitive dietary restraint and body fatness indexes. These results are also concordant with some studies (6, 9, 11, 12, 13, 14). However, negative relationships between cognitive dietary restraint and BMI have also been reported (6, 15, 19, 32). In fact, results from a weight loss intervention conducted by Foster and coworkers (19) showed a negative relationship between cognitive dietary restraint score and BMI measured before treatment. In addition, it was observed in that study (19) that a greater increase in cognitive dietary restraint was associated with a significantly larger weight loss, whereas only a weak positive relationship was noticed between decreases in disinhibition and weight loss, which led to the conclusion that an increase in cognitive dietary restraint was a stronger predictor of weight loss than a change in disinhibition (19). It is important to emphasize that the weight loss treatment in that study consisted of a very restrictive diet combined with cognitive behavioral treatment. Therefore, it is not surprising that subjects who achieved larger weight losses were those who were able to increase their restraint to follow the prescribed diet. Thus, cognitive dietary restraint seems to allow successful weight loss on a short-term period but may not necessarily promote weight maintenance on a long-term basis.

In accordance with Westenhoefer et al. (6), our results suggest that it is important to differentiate rigid restraint from flexible restraint. In fact, we observed different relationships depending on the type of cognitive dietary restraint studied. Whereas cognitive dietary restraint was not related to anthropometric variables, we noticed discordant correlation patterns for rigid and flexible restraint. Positive relationships were observed between rigid restraint and some anthropometric variables in both genders, whereas flexible restraint was negatively associated with body fat and waist circumference, but only in women. Thus, unclear relationships between cognitive dietary restraint and BMI could be explained by opposite effects of rigid and flexible restraint.

Moreover, according to Westenhoefer et al. (6), high rigid restraint is associated with higher disinhibition, suggesting that an individual following a strict and rigid diet could be more vulnerable to losing control over eating. Our results in men and women corroborate these findings. Westenhoefer et al. (6) also suggested that high flexible restraint is associated with lower disinhibition. Correlation analyses performed in our sample of men were concordant with this observation, whereas in women, no association between disinhibition and flexible restraint was found. Therefore, a more flexible restraint behavior could be a factor contributing to the avoidance of the loss of control (disinhibition), but this would be true only in men. It can also be suggested that flexible restraint could be associated with a more favorable body weight regulation, especially among obese men in whom a negative association was found between flexible restraint and BMI. Thus, these results allow us to suggest that gender may have a modulating effect on intercorrelations observed among eating behaviors.

Gender differences observed for eating behaviors in our study are concordant with other studies that have found higher scores for disinhibition and cognitive dietary restraint in women than in men (6, 9, 10, 17). Moreover, flexible restraint, rigid restraint, and the proportion of rigid to total restraint were higher in women than in men. Accordingly, Carmody et al. (9) have reported that women in our society have more often greater concern about dieting and body weight than men. In addition, classification of our subjects according to their BMI allowed us to demonstrate that gender differences in cognitive dietary restraint and disinhibition were observed for all BMI classes, except for disinhibition in obese subjects. More specifically, cognitive dietary restraint was higher in women than in men, irrespective of the obesity status. It appears that women consciously restrict their food intake more than men to control their body weight or to promote weight loss, even if BMI is in the normal range, which is concordant with previous studies (9, 10). In these regards, De Castro et al. (17) observed gender differences in factors that motivate cognitive dietary restraint. In women, cognitive dietary restraint appeared to be related to the fear of gaining weight, whereas an actual attempt to lose weight would motivate restrictive behaviors in men. Our results, which showed that both BMI categories and gender are modulating the relationships between eating behaviors and BMI, underlined the complexity of these associations. For example, although a positive relationship was found between rigid restraint and BMI among nonobese women, no such association was observed among overweight or obese women or among men. Also, in both genders, hunger was associated with BMI in overweight and obese subjects, but not among nonobese individuals. Because of the cross-sectional nature of these associations, it is not possible to determine whether eating behaviors are causally related to BMI. However, these results may suggest that individuals could respond differently, according to their BMI and gender, to interventions aimed at modifying eating behaviors to prevent weight gain or achieve weight loss.

Validity of reported energy intake was taken into consideration in our study following principles of energy physiology proposed by Goldberg et al. (24). Even if we excluded subjects whose reported energy intakes were implausibly low compared with their energy needs, we still found a significant difference in EI/RMR according to obesity status, obese subjects having lower EI/RMR values than nonobese subjects. This could suggest that obese individuals are less physically active or have relatively lower energy needs than nonobese individuals. In addition, we cannot exclude the possibility that obese individuals are more likely to underreport their food intake to some degree, as suggested by Goris et al. (33).

Regarding dietary variables, our results suggest that men and women with a more restrictive behavior are consuming less energy and dietary fat. Moreover, a more flexible restrictive approach toward eating seems to be associated with lower reported energy intake, but only in women. As defined previously, people with high cognitive dietary restraint intentionally restrict their food intake to control body weight. Therefore, it is possible that these people are more preoccupied by their food intake and could then actually eat less food. We cannot exclude the possibility of some underreporting by restrained eaters, as suggested by Bathalon et al. (34), although we had excluded subjects with obvious signs of underreporting. In both men and women, higher scores for susceptibility to hunger and internal hunger were related to a higher reported energy intake, which is concordant with previous literature (11, 13, 16). It is interesting to notice that higher scores for disinhibition and for habitual, emotional, and situational susceptibility to disinhibition were related to higher reported energy and dietary fat intakes, but only in women. These results suggest that gender may modulate the association between disinhibition and food intake. Finally, because the potential confounding effect of BMI was removed from these correlation analyses, significant positive associations observed could not be attributed to the obesity status.

In conclusion, we found significant gender differences in cognitive dietary restraint and disinhibition scores and a modulating effect of gender on the associations between eating behaviors and anthropometric profile. We also found that obese subjects had higher disinhibition and susceptibility to hunger but similar cognitive dietary restraint scores than nonobese individuals. Therefore, it appears that a higher cognitive dietary restraint is not associated with lower body weight, although a trend for such an association was found among obese men. Our results also suggest that it is important to differentiate rigid restraint from flexible restraint because unclear relationships between cognitive dietary restraint and BMI could be explained by opposite effects of rigid and flexible restraint. These findings provide further support to the concept that any intervention focusing exclusively on cognitive dietary restraint, especially rigid restraint, may not be optimal for the long-term control of body weight, particularly in women. However, in both genders, it remains to be established how eating behaviors are causally involved in the relationships observed. Further studies are needed to investigate the clinical efficiency of weight management programs focusing on the ability to cope with disinhibition and susceptibility to hunger rather than solely on dietary restraint.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

V. P. is a recipient of a studentship from Fonds de la recherche en santé du Québec (FRSQ-FCAR-Santé), and S. L. is a research scholar from the Fonds de la Recherche en Santé du Québec. This work was supported by the Canadian Institutes of Health Research (MGC-15187). The authors express their gratitude to the subjects for their excellent collaboration and the staff of the Physical Activity Sciences Laboratory for their contribution to this study. We especially thank Lucie Allard, Suzanne Brulotte, Lyne Bargone, Guy Fournier, Henri Bessette, and Claude Leblanc for their help in the collection and analysis of the data.

Footnotes
  • 1

    Nonstandard abbreviations: TFEQ, Three-Factor Eating Questionnaire; QFS, Québec Family Study; BMR, basal metabolic rate; RMR, resting metabolic rate; VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue; EI/RMR, energy intake expressed as a multiple of resting metabolic rate.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  • 1
    Ledoux, M., Rivard, M. (1998) Poids corporel. Institut de la Statistique du Québec, Enquête Sociale et de Santé 1998 187199. 2nd ed., Les Publications du Québec Sainte-Foy, Canada.
  • 2
    Statistiques Canada Indice de Masse Corporelle (IMC-Norme Canadienne) http:www.statcan.ca (accessed 2002).
  • 3
    Flegal, K. M., Carroll, M. D., Ogden, C. L., Johnson, C. L. (2002) Prevalence and trends in obesity among US adults, 1999–2000. JAMA 288: 17231727.
  • 4
    Golay, A., Hagon, I., Painot, D., et al. (1997) Personalities and alimentary behaviors in obese patients. Patient Educ Couns 31: 103112.
  • 5
    Stunkard, A. J., Messick, S. (1985) The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res 29: 7183.
  • 6
    Westenhoefer, J., Stunkard, A. J., Pudel, V. (1999) Validation of the flexible and rigid control dimensions of dietary restraint. Int J Eat Disord 26: 5364.
  • 7
    Bond, M. J., McDowell, A. J., Wilkinson, J. Y. (2001) The measurement of dietary restraint, disinhibition and hunger: an examination of the factor structure of the Three Factor Eating Questionnaire (TFEQ). Int J Obes Relat Metab Disord 25: 900906.
  • 8
    Mela, D. J. (1996) Eating behaviour, food preferences and dietary intake in relation to obesity and body-weight status. Proc Nutr Soc 55: 803816.
  • 9
    Carmody, T. P., Brunner, R. L., St Jeor, S. T. (1995) Dietary helplessness and disinhibition in weight cyclers and maintainers. Int J Eat Disord 18: 247256.
  • 10
    Neumark-Sztainer, D., Sherwood, N. E., French, S. A., Jeffery, R. W. (1999) Weight control behaviors among adult men and women: cause for concern? Obes Res 7: 179188.
  • 11
    Lawson, O. J., Williamson, D. A., Champagne, C. M., et al. (1995) The association of body weight, dietary intake, and energy expenditure with dietary restraint and disinhibition. Obes Res 3: 153161.
  • 12
    Hays, N. P., Bathalon, G. P., McCrory, M. A., Roubenoff, R., Lipman, R., Roberts, S. B. (2002) Eating behavior correlates of adult weight gain and obesity in healthy women aged 55–65 y. Am J Clin Nutr 75: 476483.
  • 13
    Lindroos, A. K., Lissner, L., Mathiassen, M. E., et al. (1997) Dietary intake in relation to restrained eating, disinhibition, and hunger in obese and nonobese Swedish women. Obes Res 5: 175182.
  • 14
    Karlsson, J., Hallgren, P., Kral, J., Lindroos, A. K., Sjöström, L., Sullivan, M. (1994) Predictors and effects of long-term dieting on mental well-being and weight loss in obese women. Appetite 23: 1526.
  • 15
    Boschi, V., Iorio, D., Margiotta, N., D'Orsi, P., Falconi, C. (2001) The three-factor eating questionnaire in the evaluation of eating behaviour in subjects seeking participation in a dietotherapy programme. Ann Nutr Metab 45: 7277.
  • 16
    Keim, N. L., Canty, D. J., Barbieri, T. F., Wu, M. M. (1996) Effect of exercise and dietary restraint on energy intake of reduced-obese women. Appetite 26: 5570.
  • 17
    de Castro, J. M. (1995) The relationship of cognitive restraint to the spontaneous food and fluid intake of free-living humans. Physiol Behav 57: 287295.
  • 18
    LaPorte, D. J., Stunkard, A. J. (1990) Predicting attrition and adherence to a very low calorie diet: a prospective investigation of the eating inventory. Int J Obes 14: 197206.
  • 19
    Foster, G. D., Wadden, T. A., Swain, R. M., Stunkard, A. J., Platte, P., Vogt, R. A. (1998) The Eating Inventory in obese women: clinical correlates and relationship to weight loss. Int J Obes Relat Metab Disord 22: 778785.
  • 20
    Bouchard, C. (1996) Genetic and body fat content. Prog Obes Res 3341.
  • 21
    Lluch, A. (1995) Identification des Conduites Alimentaires par Approches Nutritionnelles et Psychométriques: Implications Thérapeutiques et Préventives dans l'Obésité Humaines Université Henri Poincaré, Nancy I Nancy Cedex, France.
  • 22
    Laessle, R. G., Tuschl, R. J., Kotthaus, B. C., Pirke, K. M. (1989) A comparison of the validity of three scales for the assessment of dietary restraint. J Abnorm Psychol 98: 504507.
  • 23
    Health and Welfare Canada (1991) Canadian Nutrient File Government of Canada Ottawa, Ontario, Canada.
  • 24
    Goldberg, G. R., Black, A. E., Jebb, S. A., et al. (1991) Critical evaluation of energy intake data using fundamental principles of energy physiology: I. Derivation of cut-off limits to identify under-recording. Eur J Clin Nutr 45: 569581.
  • 25
    Horgan, G. W., Stubbs, J. (2003) Predicting basal metabolic rate in the obese is difficult. Eur J Clin Nutr 57: 335340.
  • 26
    The Airlie (VA) Consensus Conference (1988) Standardization of anthropometric measurements. Human Kinetics Publishers Champaign, IL.
  • 27
    Behnke, A. R., Wilmore, J. H. (1974) Evaluation and regulation of body build and composition. Behnke, AR eds. Evaluation and Regulation of Body Build and Composition 2037. Prentice-Hall Englewood Cliffs, New Jersey.
  • 28
    Meneely, G. R., Kaltreider, N. L. (1949) Volume of the lung determined by helium dilution. J Clin Invest 28: 129139.
  • 29
    Siri, W. E. (1956) The gross composition of the body. Adv Biol Med Phys 4: 239280.
  • 30
    Ferland, M., Despres, J. P., Tremblay, A., et al. (1989) Assessment of adipose tissue distribution by computed axial tomography in obese women: association with body density and anthropometric measurements. Br J Nutr 61: 139148.
  • 31
    Gendall, K. A., Joyce, P. R., Sullivan, P. F., Bulik, C. M. (1998) Personality and dimensions of dietary restraint. Int J Eat Disord 24: 371379.
  • 32
    Williamson, D. A., Lawson, O. J., Brooks, E. R., et al. (1995) Association of body mass with dietary restraint and disinhibition. Appetite 25: 3141.
  • 33
    Goris, A. H., Westerterp-Plantenga, M. S., Westerterp, K. R. (2000) Undereating and underrecording of habitual food intake in obese men: selective underreporting of fat intake. Am J Clin Nutr 71: 130134.
  • 34
    Bathalon, G. P., Tucker, K. L., Hays, N. P., et al. (2000) Psychological measures of eating behavior and the accuracy of 3 common dietary assessment methods in healthy postmenopausal women. Am J Clin Nutr 71: 739745.