Download Eating attitudes of anorexia nervosa, bulimia nervosa, binge

Physiology & Behavior 131 (2014) 99–104
Contents lists available at ScienceDirect
Physiology & Behavior
journal homepage: www.elsevier.com/locate/phb
Eating attitudes of anorexia nervosa, bulimia nervosa, binge eating
disorder and obesity without eating disorder female patients:
differences and similarities
M.S. Alvarenga a,b,⁎, P. Koritar a,b, F. Pisciolaro b, M. Mancini c, T.A. Cordás b, F.B. Scagliusi d
a
Department of Nutrition, Public Health School, University of Sao Paulo, Avenida Doutor Arnaldo, 715, São Paulo/SP 01246–904, Brazil
AMBULIM, Eating Disorders Unit of Clinics Hospital, Department of Psychiatry, Institute of Psychiatry, University of Sao Paulo, R. Dr. Ovídeo Pires de Campos, 785, São Paulo/SP 01060–970, Brazil
c
Obesity and Metabolic Syndrome/Endocrinology and Metabology Service of Clinics Hospital, University of Sao Paulo, R. Dr. Ovídeo Pires de Campos, 225, 7° Andar, Cerqueira César, São Paulo/SP
05403–010, Brazil
d
Federal University of São Paulo, Campus Baixada Santista, Av. Ana Costa, 95, Santos/SP 11060–001, Brazil
b
H I G H L I G H T S
•
•
•
•
Disordered eating attitude scale was able to distinguish eating disorder patients.
Similarities and differences are highlighted by means of relationship with food.
Anorexia and bulimia nervosa patients presented more dysfunctional eating attitudes.
Obese and binge eating disorder patients presented interesting differences.
a r t i c l e
i n f o
Article history:
Received 19 January 2014
Received in revised form 28 March 2014
Accepted 15 April 2014
Available online 24 April 2014
Keywords:
eating attitude
anorexia nervosa
bulimia nervosa
binge eating disorder
obesity
a b s t r a c t
The objective was to compare eating attitudes, conceptualized as beliefs, thoughts, feelings, behaviors and relationship with food, of anorexia nervosa (AN), bulimia nervosa (BN) and binge eating disorder (BED) patients
and a group of obese (OBS) without eating disorders (ED). Female patients from an Eating Disorder (ED) Unit
with AN (n = 42), BN (n = 52) and BED (n = 53) and from an obesity service (n = 37) in Brazil answered
the Disordered Eating Attitude Scale (DEAS) which evaluate eating attitudes with 5 subscales: relationship
with food, concerns about food and weight gain, restrictive and compensatory practices, feelings toward eating,
and idea of normal eating. OBS patients were recruited among those without ED symptoms according to the
Binge Eating Scale and the Questionnaire on Eating and Weight Patterns. ANOVA was used to compare
body mass index and age between groups. Bonferroni test was used to analyze multiple comparisons among
groups. AN and BN patients presented more dysfunctional eating attitudes and OBS patients less dysfunctional
(p b 0.001). For DEAS total score, AN and BN patients were similar and all other were different (p b 0.001).
Similarities suggested between BN and BED were true just for the “Relationship with food” and “Idea of normal
eating.” BED patients were worst than OBS for “Relationship with food” and as dysfunctional as AN patients —
besides their behavior could be considered the opposite. Differences and similarities support a therapeutic individualized approach for ED and obese patients, call attention for the theoretical differences between obesity and
ED, and suggest more research focused on eating attitudes.
© 2014 Elsevier Inc. All rights reserved.
1. Introduction
In addition to the characteristic symptoms of the syndrome, eating
disorder (ED) patients have a series of inadequate and dysfunctional attitudes toward eating with a complex relationship with food that is
marked by anxiety, anger, fear, and guilt, among other factors [1–3].
⁎ Corresponding author at: Avenida Doutor Arnaldo, 715, São Paulo/SP 01246–904,
Brazil. Tel.: +55 11 3061 7803/99196 1994.
E-mail addresses: [email protected], [email protected] (M.S. Alvarenga).
http://dx.doi.org/10.1016/j.physbeh.2014.04.032
0031-9384/© 2014 Elsevier Inc. All rights reserved.
Eating attitudes consist of beliefs, thoughts, feelings, behaviors
and relationship with food [4], and understanding these attitudes
may help to understand food choices and to target nutritional
counseling. Specifically for ED patients, the understanding of eating attitudes will help to elucidate clinical features and are good
predictors of food intake [5]. Moreover, changing disordered attitudes is important for the successful treatment of ED [6]. It is
thought that eating attitudes have implications in both treatment
and achieving greater knowledge of clinical variability in ED
patients [7].
100
M.S. Alvarenga et al. / Physiology & Behavior 131 (2014) 99–104
For anorexia nervosa (AN) and bulimia nervosa (BN), several typical
eating attitudes have been generally described: difficulty with food
choices and eating with company, dichotomy classification of food
(good or bad, safe or dangerous), incompetence in dealing with meals,
false beliefs about nutrition, angry at feeling hungry, and use of food
to address emotions [7]. These issues may be common to both AN and
BN because both disorders share certain attitudes and a higher percentage of AN patients evolve to BN [1].
Eating attitudes, using the aforementioned definition, for binge eating disorder (BED) and obese patients have not been deeply discussed.
For BED, it has been reported that they demonstrate chaotic eating
habits, high levels of loss in control and exhibit negative affect [8,9]. It
has been confirmed that most overweight and obese patients do not
overeat in any distinctive pattern [10]. Thus, it is more difficult to define
a profile of eating attitudes for obesity because obese individuals have
many different profiles; it is not possible to define an “obese personality”
because obesity is notably heterogeneous.
Nevertheless, obese patients may also have dysfunctional eating attitudes due to their eating and weight history, beliefs and cognitions incorporated during dietary practices and dietary behavioral change
efforts in trying to lose weight. However, most studies on eating behaviors in obese patients have identified it as being “unhealthy,” “uncontrolled,” “disordered,” “disinhibited,” or “restrained,” and does not
provide insight into the specific eating attitudes related to obesity
[11]. Patients may also present false nutrition beliefs; “all or nothing”
thoughts are common, which result in the lack of control behaviors, depending on whether their beliefs are right or wrong in relationship to
their choices. Frequent feelings include guilt and failure related to
their eating practices. It has been acknowledge that characteristics related to executive function, namely impulsivity and reduced decision making abilities, could result in inadequate self-control for obese patients
[12].
Studies aimed to describe eating attitude characteristics have focused on eating intake or choice [13]. Previous studies have predominantly described the eating attitudes of ED and obese patients using
scales, such as the Eating Attitude Test — EAT [14], the Eating Disorder
Examination [15] or the Eating Disorder Inventory [16]. Although useful
and well-developed psychometrically, these scales have limited scope,
tending to measure attitudes that focus on the symptoms of eating disorders, as opposed to the relationship with food in general [4].
Comparisons of ED groups and obesity are very heterogeneous in the
literature: they do not evaluate the same groups, and some studies are
quite old, and discuss mostly aspects of the disease (e.g., presentation,
personality, comorbidities). These studies do not focus on eating attitudes, or at least not in a broad view [17–19]. More studies have compared BED and non-BED obese patients, but these studies considered
other issues, such as appetite-related substances, cognitive function, energy intake, and personality [20–23].
The comparison of the eating attitudes of AN, BN, BED and obese patients could be considered of interest, both clinically and theoretically.
In this context, the objective of the present study was to evaluate eating
attitudes, focusing on the beliefs, thoughts, feelings, behaviors and relationship with food, exhibited by ED and obese patients and to compare
potential similarities and differences.
2. Materials and methods
2.1. Sample
Female eating disorder patients were evaluated from a group receiving multiprofessional treatment at the Eating Disorders Unit of University of Sao Paulo (USP), Brazil. Anorexia nervosa, bulimia nervosa and
binge eating disorder patients who were accepted at this unit of treatment during 2009 until 2012 were recruited to participate in
this study. Obese women – diagnosed by Body Mass Index (BMI) ≥
30 kg/m2 – patients seeking multiprofessional treatment at the Obesity
Outpatient of University of Sao Paulo (USP), Brazil were invited to participate in this study during the 2012 recruitment.
To participate, the ED patients must meet the following inclusion
criteria: age between 18 and 45 years old, literate, and do not have specialized ED treatment for at least one month. Obese patients (OBSpat)
must meet the following criteria: age between 18 and 45 years old, literate, do not use specialized medication with the objective to lose
weight for at least one month, and do not have BED symptoms. To exclude those with BED symptoms in the obese group, participants were
asked to complete the Binge Eating Scale – BES [24] and the Questionnaire about Eating and Weight – QEW [25]. From the 73 patients who
were contacted, 36 patients received a score ≥ 17 for BES and/or
presented BED symptoms according to the QEW.
2.2. Procedures
ED was diagnosed using criteria of the Diagnostic and Statistical
Manual of Mental Disorders (DSM-IV) according to a psychiatric interview. On the first day of treatment, the patients completed the study instrument and had their weight and height measured by a dietitian to
calculate their BMI. The procedure was performed in the waiting room
of the ED Unit.
Obese patients were recruited during a screening or while they were
waiting for their first visit to the Obesity Outpatient Clinic. Their weight
and height were measured on the same day by a nurse and the study instruments were completed in the outpatient dependencies.
2.3. Instruments
Eating attitudes were evaluated using the Disordered Eating Attitude
Scale — DEAS. The DEAS consists of 25 questions, which are scored on
the basis of the Likert scale, ranging from 37 to 185. The DEAS includes
five subscales, known as:
Subscale 1 — Relationship with food: evaluates attitudes related to
the ways that individuals address food in terms of food control, food
refusal, guilt, anger, desire and shame (score ranging 12–60); Subscale
2 — Concerns about eating and body weight gain: evaluates concerns regarding calories, intake control, obsessive thoughts about food and
weight gain (score ranging 4–20); Subscale 3 — Restrictive and compensatory practices: evaluates the restriction of food and calories, and
attitudes aimed to compensate large or uncontrolled food intake
(score ranging 4–20); Subscale 4 — Feelings toward eating: evaluates
feelings concerning pleasure and food memories and how normal one
feels to eat (score ranging 3–15); and Subscale 5 — Idea of normal eating: evaluates rigid nutrition concepts and beliefs (score ranging
14–70). The higher the score in the scale, the more dysfunctional is
the attitude [4].
The DEAS was developed with Brazilian women college students
and found good internal consistency (Cronbach's Alpha 0.75), and convergent and known-group validity [4] — which means it has acceptable
reliability and is valid. For the present groups evaluated, the Cronbach's
Alpha were: 0.84 for AN; 0.76 for BN; 0.80 for BED and 0.66 for OBS —
indicating good reliability for ED groups and acceptable for obesity. Besides that, a comparison of Alphas in these different samples, performed
by Hakstian-Whalen test [26], showed no significant difference (p =
0.14; M = 5.38).
The scale was also validated for women adult individuals in English,
Spanish and Japanese languages [27–29] and was thought to be useful
for studying eating aspects in different populations and highlighting differences in attitudes among diverse groups and clinical populations.
Regarding the scales used to evaluate BED symptoms; BES was developed by Gormally et al. (1982) [24] and was adapted for the
Brazilian context [30]. It is a 16-item scale, which evaluates the severity
of binge eating and provides classification as following: absence of regular binge eating (score ≤ 17), moderate binge eating presence (score
between 18 and 26), severe binge eating presence (score ≥ 27). The
M.S. Alvarenga et al. / Physiology & Behavior 131 (2014) 99–104
QEW [25] was validated for BED screening in Brazil [31], it consists of 26
questions with a template to evaluate bulimia nervosa (purging or not
purging subtypes) and BED symptoms — according to answers for
specific questions.
2.4. Ethics
The study protocol was approved by the ethics committee of the
Public Health School of the University of Sao Paulo, and had the approval
of the direction of the Eating Disorders Unit and Obesity Outpatient
Unit. All participants of this study provided their written consent.
2.5. Statistical analysis
Statistical analyses were performed using SPSS 18.0 (Statistical Package for Social Science Inc., Chicago, Illinois, USA). The significance level
adopted was 0.05. Normality was tested using the Shapiro-Wilk test
and non-normal variables (DEAS scores, age, BMI) were standardized
using the Z-score.
ANOVA was used to compare BMI and age between groups. Multiple
comparisons were performed using Bonferroni's test. A general linear
model (GLM) analysis was used to evaluate DEAS total and subscale
scores among groups, controlling for BMI and age. The confidence interval coefficient was analyzed and pairwise comparisons between groups
were performed. Effect size (partial Eta square) and observed power for
comparisons are also presented.
3. Results
Female ED patients from the Eating Disorders Unit who participated
in this study consisted of 42 subjects with anorexia nervosa (ANpat), 52
subjects with bulimia nervosa (BNpat) and 53 subjects with binge eating disorder (BEDpat). Obese patients without BED symptoms from
the Obesity Outpatient (OBSpat) consisted of 37 subjects. The age and
BMI of the patients are shown in Table 1.
As expected, obese and BED patients were older than the AN and BN
subjects, but not different between them. BN patients were older than
AN patients — although the comparison found a p = 0.051, the effect
size was d = −0.599 (Cohen's d), which attested for the difference.
Regarding BMI, AN patients, who are compatible with their disorder,
had the lowest values, BNpat intermediate, and BED and obese patients
had highest values with no difference among them.
The DEAS total and subscale scores for all of the different groups are
shown in Table 2.
Table 1
Means and 95% confidence intervals (CI) of age and body mass index of anorexia nervosa
(AN), bulimia nervosa (BN), binge eating disorder (BED) and obese female patients
without BED symptoms (OBS) (n = 184).
DIAGNOSTIC
AN (n = 42)
BN (n = 52)
BED (n = 53)
OBS (n = 37)
Age
Body Mass Index
Mean (95% CI)
Mean (95% CI)
24.9 (23.1–26.8)a
29.5 (27.0–31.9)b
36.45 (34.0–39.0)c
39.14 (36.4–41.9)d
p value b 0.001
16.73 (16.2–17.3)e
25.07 (23.5–26.6)f
38.72 (36.9–40.6)g
38.55 (36.7–40.4)h
p value b 0.001
a
Different from BN (p = 0.051*), BED (p b 0.001) and OBS (p b 0.001); *Effect size
(Cohen's d) d = −0.599, which attested for the difference.
b
Different from AN (p = 0.051*), BED (p b 0.001) and OBS (p b 0.001); *Effect size
(Cohen's d) d = −0.599, which attested for the difference.
c
Different from AN and BN (p b 0.001).
d
Different from AN and BN (p b 0.001).
e
Different from BN (p b 0.001), BED (p b 0.001) and OBS (p b 0.001).
f
Different from AN (p b 0.001), BED (p b 0.001) and OBS (p b 0.001).
g
Different from AN (p b 0.001) and BN (p b 0.001).
h
Different from AN (p b 0.001) and BN (p b 0.001).
101
For the DEAS total score, higher values were found for AN and BN patients, with no difference among them. ANpat also had higher scores
than BED and OBS. BNpat scored higher than BEDpat and obese patients.
BEDpat had higher DEAS total score than OBSpat.
Regarding subscale 1 – Relationship with food – the highest score
was found to BNpat group and the lowest was found to OBSpat group,
with significant differences between ANpat and BNpat, ANpat and
OBSpat, BNpat and OBSpat, BNpat and BEDpat, and also for BED and
OBS.
Analysis of subscale 2 – Concerns about eating and body weight
gain – found that ANpat and BNpat had the highest and OBSpat
lowest scores. Significant differences were observed between AN
and OBS, BN and BED, BN and OBS.
Subscale 3 – Restrictive and compensatory practices – showed the
same pattern: ANpat and BNpat had the highest and OBSpat lowest
values. Significant differences were found between ANpat and BEDpat,
ANpat and OBSpat, BNpat and BEDpat, BNpat and OBSpat, BEDpat and
OBSpat.
For subscale 4 – Feelings toward eating – ANpat and BNpat had
highest scores, and BEDpat and OBSpat lowest ones. AN was significant
different from BED and OBS, and BN was also significant different from
BED and OBS.
Lastly, for subscale 5 – Idea of normal eating – similar scores for
groups were observed, with higher values for ANpat and BNpat and
lowest for BEDpat and OBSpat. Significant differences were found just
between AN and BED, AN and OBS. However, the effect size for between
comparisons regarding subscale 5 was 0.08 – below a minimum of
0.15 – indicates that a type 1 error could have been present, suggesting that the magnitude of the difference was not real.
4. Discussion
This study performed a comparison of disordered eating attitudes
among women with different ED diagnosis and obese patients and
found interesting similarities and discrepancies. To the best of our
knowledge, this is the first study that reports such an evaluation,
which embodies eating attitudes as beliefs, thoughts, feelings, behaviors
and relationship with food.
Regarding age, higher values for BED and OBS patients were observed, which was consistent with the profile of adult clinical samples
elsewhere [32,33]. The older BN patients (even not showing significant
difference) compared with AN were also consistent with the knowledge
of a higher age of incidence for BN [1] — even studying patients from an
adult service.
As expected, the BMI was lower for AN, intermediate for BN and
higher (and similar) for BED and OBS patients. It is know that usually
BED patients have a more superior BMI than BN [3,34], and have the
risk of putting on weight during the course of disorder — which can
evolve into an obese condition [10]. Even thought BMI were different
among groups, the covariance analysis of the eating attitude (DEAS
total and subscales) was performed controlling for BMI and age.
AN and BN are considered the “most serious” ED, with higher effects
for patients global and clinical health [1,3], and analysis of data on eating
attitudes revealed more dysfunctional eating attitudes as a whole in
these two subgroups. BNpat presents psychological characteristics,
such as impulsivity and anxiety symptoms, which translate into chaotic
eating patterns and attitudes [7,35]. Furthermore, AN patients experience a higher necessity to control their environment, inflexible
thoughts, perfectionism and limited social spontaneity [12,35] —
which is reflected in their eating attitudes.
Despite issues related to eating behavior in obesity, OBS patients had
the lowest value for the DEAS total scale, i.e., less dysfunctional eating
attitudes. This result was important even more when they were differentiated from BED patients. Specific studies evaluating eating attitudes,
which use the presented definition of this construct in BED were not
found; clinical observations demonstrate lower levels of distorted
102
M.S. Alvarenga et al. / Physiology & Behavior 131 (2014) 99–104
Table 2
Means and 95% confidence intervals (CI) of Disordered Eating Attitude Scale (DEAS) total and subscale scores for anorexia nervosa (AN), bulimia nervosa (BN), binge eating disorder (BED)
and obese female patients without BED symptoms (OBS) (n = 184).
Diagnostic
AN
(n =
BN
(n =
BED
(n =
OBS
(n =
DEAS total score
Total
Subscale 1
Subscale 2
Subscale 3
Subscale 4
Subscale 5
Mean (95% CI)
Mean (95% CI)
Mean (95% CI)
Mean (95% CI)
Mean (95% CI)
Mean (95% CI)
42) 110.6 (97.4–121.8)a
36.45 (31.7–41.9)e
13.24 (11.5–16.1)i
14.71 (13.3–17.5)m
9.57 (7.8–11.2)q
36.67 (30.7–40.0)u
52) 119.1 (110.2–125.8)b
46.83 (43.7–50.2)f
14.71 (13.4–16.4)j
15.50 (14.4–17.1)n
9.75 (8.6–10.8)r
32.21 (28.7–34.7)v
53) 89.7 (81.0–99.8)c
39.57 (35.4–43.3)g
10.42 (8.3–11.8)k
8.83 (6.7–10.0)o
4.96 (3.7–6.3)s
25.96 (23.1–30.2)x
37) 65.1 (54.5–75.5)d
p value b 0.001 Eta
square = .322; Observed
power = 100%
19.92 (15.2–24.1)h
p value b 0.001 Eta
square = .419; Observed
power = 100%
8.27 (5.9–9.9)l
p value b 0.001 Eta
square = .176; Observed
power = 100%
5.95 (3.6–7.3)p
p value b 0.001 Eta
square = .370; Observed
power = 100%
4.19 (2.8–5.7)t
p value b 0.001 Eta
square = .275; Observed
power = 100%
26.68 (23.8–31.9)z
p value = 0.002 Eta
square = .081; Observed
power = 92%
Subscale 1 = Relationship with food; Subscale 2 = Concerns about eating and body weight; Subscale 3 = Restrictive and compensatory practices; Subscale 4 = Feeling toward eating;
Subscale 5 = Idea of normal eating.
a
Different from BN (p = 0.618), BED (p = 0.005) and OBS (p b 0.001).
b
Different from BED (p b 0.001) and OBS (p b 0.001).
c
Different from OBS (p b 0.001).
d
Different from AN (p b 0.001), BN (p b 0.001) and BED (p b 0.001).
e
Different from BN (p b 0.001) and OBS (p b 0.001).
f
Different from OBS (p b 0.001) and BED (p = 0.006).
g
Different from OBS (p b 0.001).
h
Different from AN (p b 0.001), BN (p b 0.001) and BED (p b 0.001).
i
Different from OBS (p b 0.01).
j
Different from BED (p b 0.001) and OBS (p b 0.001).
k
Different from BN (p b 0.001).
l
Different from AN (p b 0.001) and BN (p b 0.001).
m
Different from BED (p b 0.001) and OBS (p b 0.001).
n
Different from BED (p b 0.001) and OBS (p b 0.001).
o
Different from OBS (p = 0.017).
p
Different from AN (p b 0.001), BN (p b 0.001) and BED (p = 0.017).
q
Different from BED (p b 0.001) and OBS (p b 0.001).
r
Different from BED (p b 0.001) and OBS (p b 0.001).
s
Different from AN (p b 0.001) and BN (p b 0.001).
t
Different from AN (p b 0.001) and BN (p b 0.001).
u
Different for BED (p = 0.001*) and OBS (p = 0.022*) *Effect size (Cohen's d), for between comparisons = 0.08, indicates type 1 error.
beliefs, thoughts and feelings compared to AN and BN patients. According to Stunkard (2011) [10], compared with non-BED obese, BED patients experience greater psychopathology, and they also report less
exposure to risk factors for general psychopathology compared to individuals with BN. Our results were consistent with these reports.
Importantly, obesity is also related to feelings of guilt, anxiety and is
associated with the lack of strategies to cope with stress in an effective
manner [36]. All these issues could affect eating attitudes, but the
DEAS total score and 2 subscales (“Relationship with food” and “Restrictive and compensatory practices”) were different for OBSpat and
BEDpat. These results indicate that even with similar psychological
characteristics, an ED associated with obesity (as is the case for BED)
presents much more disturbed and socially impaired eating attitudes.
Recently, Pollert (2013) [9] asserted that the perception of a loss in control is an inherent component of BED compared with non-BED participants, independently of the effects of caloric intake and affect.
Carter and Jansen (2012) [11] further reported that one should be
careful to not assume that concepts inherited from eating disorders
will directly translate to obesity. Therefore, another instrument to evaluate disordered feelings, thoughts and behaviors in obesity could be
required.
When specifically considering the subscales, a more dysfunctional
“Relationship with food” was found for BN patients — which was different from all the others, including AN. It may be affirmed that AN patients
achieve the goal proposed by our culture, which suppresses women
hunger and makes women ashamed of their appetite and needs [37].
Consistent with this thought, a woman with AN could not see her relationship with food dysfunctional or disordered; she feels that she overcame food. Thus, AN may give the individual control over food, whereas
BN behaves in the opposite manner. A woman with BN also believes
that she should not eat, but she blames herself because she cannot do
that, thus having feelings of failure and unsuccessfulness [37].
Our clinical observation showed that BN patients are more prone to
express their thorny relationship with food; they experience more
anger, guilt, ambivalence and incompetence for food. This impaired relationship is thought to be a main challenge of nutritional therapy for
BNpat [4].
Surprisingly, for “Relationship with food,” BEDpat and ANpat
showed similar scores, although the BNpat was different from BEDpat,
and the BEDpat was different from the OBSpat. Thus, despite a high seriousness related to the AN consequences, the BEDpat showed no difference in the effect on “Relationship with food” compared with ANpat.
Nevertheless, they could still have a desire to control their eating similar
to ANpat, but are not able to act on this desire due to an increase in impulsivity, which can be related to their emotional vulnerability and deficiency in the abilities to modulate negative emotions in a functional
manner [9]. This result may indicate a similar troubled relationship
with food, which could result in completely distinct nutritional status
results and consequences.
Similar considerations may be entertained for the results on “Concerns about eating and body weight gain,” which were similar for
ANpat and BNpat as well as for ANpat and BEDpat. BNpat showed the
highest level of concern regarding calories, food intake control and obsessive thoughts regarding eating and weight, but ANpat and BEDpat
were similar. Thus, a comparable intensity of worries with regard to
food and weight could also result in different results: AN patients
showed a higher ability to respond to this with serious eating restrictions, whereas BEDpat and OBSpat are not “successful” in their attempts
M.S. Alvarenga et al. / Physiology & Behavior 131 (2014) 99–104
to lose weight despite as the concern of AN patients about putting on
weight. However, the BED and OBS patients also showed a similar
level of concern in the present evaluation. This could be explained by
the fact that stigmatization and making themselves responsible by
their condition regarding weight is observed in BED and obesity – and
not in other ED – because they attributed their condition to a lack of
self-control [38].
These results of similarity between ANpat and BEDpat were consistent with the knowledge that ED and obesity may share some biological
and intermediate neurocognitive phenotypes, such as cognitive vulnerabilities. ED and obese patients may share specific neurobiological correlates that are related to the food reward system and may only vary
on the frequency or severity of these traits [12]. These results seem rationale when considering that BED may be thought of as a complex
presentation of obesity, with more impaired cognitive dysfunction.
Consistent with their ED symptoms, AN and BN patients were not
different for “Restrictive and compensatory practices.” Furthermore,
ANpat and BNpat exhibited more of these behaviors compared with
BEDpat [3]. In addition, BEDpat had higher scores for this subscale
than OBSpat. For definition, there are no inappropriate compensatory
behaviors to counteract the potential effects of binging on weigh BED,
and thus, it could be concluded that the result of higher “restrictive
and compensatory practices” for BED indicated they have more restrictive practices than OBS patients. Nevertheless, one of the DEAS question
included in this subscale (“When you eat more than usual, what is your
behavior afterwards?”), has as option for the potential to go on a diet to
compensate [4], and even not precisely considered as a “inappropriate
compensatory behavior,” it appears that BED patients could have more
compensatory ideas and behaviors (such as diet, to eat less after exaggeration in eating) than OBS individuals.
For “Feeling toward eating,” the AN and BN patients' scores were
similar and higher than BED and OBS individuals, which were also
alike. These results indicated that the AN and BN patients feeling less
normal and had less pleasure in relationship to food. BED and OBS
patients appeared to be less dysfunctional with reference to this
feeling – which was analogous between them – using food more as
a source of pleasure than punishment.
Considering the higher score for BN and AN patients for “Relationship with food” and “Feeling towards eating” compared with BED and
OBS patients, it is interesting to speculate what came first: did these
AN and BN women have a less positive feeling about eating, hard experiences with food that put them at risk for ED; or were they so submerged in the ED, that food became the opposite of something that
was natural, good and pleasant? For BED and obesity patients eating
also involves social pressures and guilt, but the effect on their relationship and feelings with food appeared to be lower than AN and BN.
Rigid concepts and beliefs about eating, as evaluated by subscale 5,
showed a difference only for AN compared with BED and OBS. However,
the effect size showed that a real difference did not exist because the
magnitude was minimal. This result makes sense if we remember that
the idea about “normal eating” is a cultural concept, which is currently
focused on the condemnatory values and messages that were disseminated by the media, families, peers and health professionals [39]. Thus,
it is reasonable that all of the groups presented similar scores and
showed no difference.
It is difficult to compare the presented results with the literature because similar studies (a comparison of AN, BN, BED and OBS regarding
disordered eating attitudes) were not found. Moreover, studies make
comparisons using heterogeneous samples, diverse instruments and
the construct of eating attitudes are not well defined — and the adopted
definitions tend to differ among them [8,12,18,20,22].
Nevertheless, despite the instrument, purpose and sample, comparisons between BN and BED patients (including other ED groups and
obese individuals) normally showed that the BN group presented increased psychopathology [17–19,22]. Using Eating Disorder Inventory
and Eating Disorder Questionnaire, Raymond et al. (1995) [19] found
103
that BNpat experienced more eating- and weight-related pathology
compared to BED patients. In addition, BEDpat were less anxious and
less worried about their eating patterns than BNpat, and binge eating
was less ego dystonic in BEDpat than BNpat.
Furthermore, the result that BEDpat had more dysfunctional eating
attitudes compared with obese individuals is consistent with the finding
by Raymond (1995) [19] that BED participants were more pathologically disturbed than obese participants who did not binge eat. For example,
Marcus (1992) [17] found higher rates of eating disorder symptoms as
assessed by Eating Disorder Inventory in BEDpat compared with nonbinge eating obese patients. Moreover, Fassino (2003) [8] found higher
rates of depression and impulsivity for BED obese patients than in obese
patients without BED.
The finding of the present discrepancies and similarities has several
implications. The DEAS did not determine “levels of normality” in relationship to eating attitudes, but similar total scores for BED and OBS individuals demonstrated that obese patients have issues that are far from
the quali-quantitative aspects to be assessed in their eating behavior.
This is consistent with Carter and Jansen (2012) [11] who stated that
it is “important to be aware of the full range of eating behaviors that
are potentially relevant to obesity, since different eating behaviors
may be problematic for different people.”
However, OBS individuals had a better “relationship with food” and
lower “restrictive and compensatory practices” than BED patients. Notwithstanding, the similarity of BED and OBS with regards to the “concerns about eating and body weight” suggested that obesity per se was
not the major factor in weight gain discomfort, but that BEDpat also developed dysfunctional eating concepts that make them similar to AN patients, and may potentially involve a worse “relationship with food.”
The characteristics demonstrated for BEDpat with DEAS comparisons supported the fact that BED is not a simple behavioral subtype of
obesity, but rather an ED diagnostic, as recently published in DSM-5
[3] because of its similarities with other ED issues. This was consistent
with Schulz and Laessle (2010) [40], who found higher levels of comorbidities and emotional eating in BEDpat compared with obese patients.
The present study has important strengths, including a comparison
of all classic, and more prevalent, ED diagnoses (AN, BN and BED) and
obesity without ED symptoms. In addition, an evaluation of eating was
beyond the classical issues evaluated, such as intake or clinical pathology, but assessing beliefs, thoughts, feelings, behaviors and relationship
with food. Consistent with the thought from Walsh (2011) [41] regarding eating behavior, understanding eating attitudes may help to determine which factors contribute to the development and persistence of
eating disturbances, or to comprehend how they change with
treatment.
Nevertheless, we must consider some limitations for the present
study. The instrument used in this evaluation was a scale that translated
quantitative data. Using the DEAS, we could attribute this to subjective
variables, but may not observe or investigate the identity of the feelings,
thoughts, beliefs and behaviors in each group. Nevertheless, this focus
would be potentially used for deep interviews or participative observations, which are rare methods in ED studies — but are commonly used to
obtain clinical observations or quantitative data. We are aware that by
using a single instrument, we “framed” the individuals in previous
known attitudes. Thus, we compared if these attitudes were presented
in higher or lower intensity among groups, but we cannot say if an investigation about the attitudes characteristic from each group was
performed.
In addition, independent variables, which should predict to have an
effect on eating attitudes, such as the environment, working conditions,
educational level, genetic factors and psychological state [36] were not
evaluated. The obese group was a treatment-seeking one, and they
may be different from the non-treatment-seeking group (but ED
group was also studied among a population of a specialized treatment
center). Information of the duration of obesity was not collected or information on whether they had ever attempted to lose weight (or
104
M.S. Alvarenga et al. / Physiology & Behavior 131 (2014) 99–104
how many times), but we also did not collect the duration of the ED or
beginning of the ED symptoms. Finally, only women were included, and
thus the results were not applicable to men. However, because EDs are
mostly found in women, a sample of men would be difficult to achieve
to make comparisons. Thus, the results should be interpreted in the context of these limitations.
However, these results support a therapeutic individualized approach for ED and obese patients. In a nutritional perspective, it is necessary to focus on interventions that are consistent with the patient's
attitudes profile to maximize effectiveness. Once the treatment considers the different aspects of relationship with food, it might offer significant help in improving the recovery of ED and the healthy living
style for obese individuals [11].
5. Conclusions
It was concluded that AN and BN patients presented more dysfunctional eating attitudes, and their scores on the instrument were similar
in general; obese patients had the lowest score (e.g., less dysfunctional
eating attitudes).
“Relationship with food” was the factor that differentiated or approximated most of the groups of ED and obesity, followed by “Feelings
toward eating,” which showed that it could be the major difference
among these groups regarding disordered eating attitudes.
The similarity suggested by several studies between BN and BED
were true only for the “Relationship with food” and the “Idea of normal”
eating in this study. Consistent with this notion, similarity between BED
and OBS was not found for “Relationship with food” with BED patients
presenting a more dysfunctional one. Moreover, BED patients appeared
to have a disturbed “relationship with food” similar to AN patients — in
spite to opposing behaviors.
Differences and similarities found among groups support a therapeutic individualized approach for ED and obese patients, call attention
for the theoretical differences between obesity and ED, and suggest
more research focused on beliefs, thoughts, feelings, behaviors and relationship with food.
Conflict of interest
Authors declare no conflict of interest, financial or otherwise, related
to the submitted work.
Acknowledgements
The authors would like to thank The State of Sao Paulo Research
Foundation – FAPESP (process 06/56850-9) – for the grant awarded to
the first author that supported this research and Bernardo dos Santos
for statistical support. The last author is supported by the National
Council of Scientific and Technology Development (CNPq, process
309121/2012-4).
References
[1] American Psychiatric Association (APA). Practice Guidelines for the treatment of patients with eating disorders. Washington, DC: American Psychiatric Association;
2006.
[2] American Dietetic Association. Position of the American Dietetic Association: Nutrition Intervention in the Treatment of Eating Disorders. J Am Diet Assoc
2011;111:1236–41.
[3] American Psychiatric Association (APA). Diagnostic and Statistical Manual of Mental
Disorders5th ed. ; 2013 [Washington DC].
[4] Alvarenga MS, Scagliusi FB, Philippi ST. Development and validity of the disordered
eating attitude scale (DEAS). Percept Mot Skills 2010;110:379–95.
[5] Sunday SR, Einhorn A, Halmi KA. Relationship of perceived macronutrient and caloric content of affective cognitions about food in eating-disordered, restrained, and
unrestrained participants. Am J Clin Nutr 1992;55:362–71.
[6] Eiger MR, Christie BW, Sucher KP. Change in eating attitudes: an outcome measure
of patients with eating disorders. J Am Diet Assoc 1996;96:62–4.
[7] Alvarenga MS, Scagliusi FB, Philippi ST. Changing attitudes, beliefs and feelings towards food in bulimic patients. Arch Latinoam Nutr 2008;58(3):274–9.
[8] Fassino S, Leombruni P, Piero A, Abbate-Daga G, Rovera GG. Mood, eating attitudes
and anger in obese women with and without Binge Eating Disorder. J Psychosom
Res 2003;54:559–66.
[9] Pollert GA, Engel SG, Schreiber-Gregory DN, Crosby RD, Cao L, Wonderlich SA, et al.
The role of eating and emotion in binge eating disorder and loss of control eating. Int
J Eat Disord 2013;46(3):233–8.
[10] Stunkard AJ. Eating disorders and obesity. Psychiatr Clin North Am 2011;34:765–71.
[11] Carter FA, Jansen A. Improving psychological treatment for obesity. Which eating behaviours should we target? Appetite 2012;58:1063–9.
[12] Fagundo AB, de la Torre R, Jímenez-Murcia S, Agüera Z, Granero R, Tárrega S, et al.
Executive functions profile in extreme eating/weight conditions: from anorexia
nervosa to obesity. PLoS One 2012;7(8) [e43382 pp.].
[13] Heaner MK, Walsh BT. A history of the identification of the characteristic eating disturbances of Bulimia Nervosa, Binge Eating Disorder and Anorexia Nervosa. Appetite
2013;65:185–8.
[14] Garner DMOM, Bohr Y, Garfinkel PE. The eating attitude test: psychometric features
and clinical correlates. Psychol Med 1982;12:871–9.
[15] Cooper Z, Cooper PJ, Fairburn CG. The validity of the eating disorder examination
and its subscales. Br J Psychiatry 1989;154:807–12.
[16] Garner DM, Olmstead MP, Polivy J. The development and validation of a multidimensional eating disorder inventory for anorexia and bulimia. Int J Eat Disord
1983;2:15–34.
[17] Marcus MD, Smith D, Santelli R, Keye W. Characterization of eating disordered behavior in obese binge eaters. Int J Eat Disord 1992;12:249–55.
[18] Fichter MM, Quadflieg M, Brandl B. Comparison of binge eating disorder, bulimia
nervosa and obesity. Int J Eat Disord 1993;14:1–16.
[19] Raymond NC, Mussell MP, Mitchell JE, de Zwaan M, Crosby RD. An age-matched
comparison of participants with binge eating disorder and bulimia nervosa. Int J
Eat Disord 1995;18:135–43.
[20] Raymond NC, Bartholome LT, Lee SS, Peterson RE, Raatz SK. A comparison of energy
intake and food selection during laboratory binge eating episodes in obese women
with and without a binge eating disorder diagnosis. Int J Eat Disord 2007;40:67–71.
[21] Geliebter A, Hashim SA, Gluck ME. Appetite-related gut peptides, ghrelin, PYY, and
GLP-1 in obese women with and without binge eating disorder (BED). Physiol
Behav 2008;94:696–9.
[22] Peterson CB, Thuras P, Ackard DM, Mitchell JE, Berg K, Sandager N, et al. Personality
dimensions in BN, BED and obesity. Compr Psychiatry 2010;51:31–6.
[23] Galioto R, Spitznagel MB, Strain G, Devlin M, Cohen R, Paul R, et al. Cognitive function in morbidly obese individuals with and without binge eating disorder. Compr
Psychiatry 2012;53:490–5.
[24] Gormally J, Black S, Daston S, Rardin D. The assessment of binge eating severity
among obese persons. Addict Behav 1982;7:47–55.
[25] Spitzer RL, Devlin M, Walsh BT, Hasm D, Wing R, Marcus MD, et al. Binge eating disorder: a multisite field trial of the diagnostic criteria. Int J Eat Disord
1992;11:191–203.
[26] Hakstian AR, Whalen TE. A k-sample significance test for independent alpha coefficients. Psychometrika 1976;41:219–31.
[27] Alvarenga MS, Pereira RF, Scagliusi FB, Philippi ST, Estima CC, Croll J. Psychometric
evaluation of the Disordered Eating Attitude Scale (DEAS). English version. Appetite
2010;55:374–6.
[28] Alvarenga MS, Francischi R, Fontes F, Scagliusi FB, Philippi ST. Adaptación y
validación al Espanol del Disordered Eating Attitude Scale (DEAS). Perspectivas
Nutr Hum 2010;12:11–23.
[29] Chisuwa N, Shimai S, Haruki T, Alvarenga MS. Development of the Japanese Version
of the Disordered Eating Attitude Scale (DEAS). Sch Health 2013;9:14–22.
[30] Freitas S, Lopes CS, Coutinho W, J.C. A. Tradução e adaptação para o português da
escala de compulsão alimentar periódica. Rev Bras Psiquiatr 2001;23:215–20.
[31] Borges MBF, Morgan CM, Claudino AM, da Silveira DX. Validação da versão em
português do Questionário sobre Padrões de Alimentação e Peso: revisado
(QEWP-R) para o rastreamento do transtorno da compulsão alimentar periódica.
Rev Bras Psiquiatr 2005;27:319–22.
[32] Smink FRE, Van Hoeken D, Hoek HW. Epidemiology of eating disorders: incidence,
prevalence and mortality rates. Curr Psychiatry Rep 2012;4:406–14.
[33] Kessler RC, Berglund PA, Chiu WT, Deitz AC, Hudson JI, Shahly V, et al. The prevalence and correlates of binge eating disorder in the World Health Organization
World Mental Health Surveys. Biol Psychiatry 2013;73:904–14.
[34] Grucza RA, Przybeck TR, Cloninger CR. Prevalence and correlates of binge eating disorder in a community sample. Compr Psychiatry 2007;48:124–31.
[35] Rawal A, Park RJ, Williams MG. Rumination, experiential avoidance, and dysfunctional thinking in ED. Behav Res Ther 2010;48:851–9.
[36] Keskin G, Engin E, Dulgerler S. Eating attitude in the obese patients: the evaluation in
terms of relational factors. J Psychiatr Ment Health Nurs 2010;17:900–8.
[37] Scagliusi FB, Nakagawa KA, Campos RM, Kotait M, Fabbri A, Sato P, et al. Nutritional
knowledge, eating attitudes and chronic dietary restraint among men with eating
disorders. Appetite 2009;53:446–9.
[38] Ebneter DS, Latner JD. Stigmatizing attitudes differ across mental health disorders: a
comparison of stigma across eating disorders, obesity, and major depressive disorder. J Nerv Ment Dis 2013;201:281–5.
[39] Paquette MC. Perceptions of healthy eating: state of knowledge and research gaps.
Can J Public Health 2005;96(Suppl. 3):S15–9 [S16a-21a].
[40] Schulz S, Laessle RG. Associations of negative affect and eating behaviour in obese
women with and without binge eating disorder. Eat Weight Disord 2010;15:
e287–93.
[41] Walsh BT. The importance of eating behavior in eating disorders. Psychol Behav
2011;104:525–9.