Basic HTML Version
Porto, Portugal | September 20-22, 2012
39
eating concern to 0.70 (0.81) for shape concern. Reliability
coefficients for the subscales were generally lower for men
(0.64 to 0.84). Our mean global EDE-Q score was less than
half of the average reported in a US sample (Lavender et al.,
2010).
Discussion:
EDE-Q scores among male community
sample were significantly, and almost invariably, lower than
similarly-aged young women. The EDE-Q performed less
reliably among young men, and this is an important caveat.
Owing to the overall pattern of low item endorsement, reliance
upon this single instrument is unlikely to provide a compre-
hensive assessment of shape, weight, and eating concerns
among non-clinical, normal-weight young men.
T57
SELF-PERCEPTION OF BODY IMAGE AND RISK EATING
BEHAVIORS IN COLOMBIAN NATIVE AND AFRO-DE-
SCENDENT POPULATION
Maritza B. Rodríguez
1,2,3,4
, Juanita Gempeler
4,5,6
, Olga L
Sarmiento
7
, Daniela Cleves
8
1
M.D: Psychiatrist Javeriana University, Bogotá, Columbia,
2
MSc Epidemiologist Javeriana University, Bogotá, Columbia,
3
Associated Professor Javeriana University, Bogotá, Co-
lumbia,
4
Scientific Director Equilibrio Outpatient Program for
ED Treatment, Bogotá, Columbia,
5
Psychologist Javeriana
University, Bogota, Columbia,
6
Diploma in Behavioral Psycho-
therapy London University, London, United Kingdom,
7
Epide-
miologist Los Andes University, Bogotá, Columbia,
8
Medical
Student Los Andes University, Bogotá, Columbia
Purpose
:Examine the prevalence of the discordance (over-
estimation or sub estimation) in body image self perception
in a sample of natives and afrodescendents and compare it
with the population of all the country.Risk eating behaviors
were also compared.
Methods:
A secondary analysis of the
National Survey of Colombian Nutritional Situation ENSIN
2010 was performed:13595 people (13-64 years old) from
all stratum in rural and urban areas of Colombia. A sample
of 1332 natives and 1368 afrodescendentes was analyzed.
The discrepancy between self perception of body image and
BMI as well as risk eating behaviors were explored.
Results
:In
adolescents 10,3% (CI 95% 7,6-13,9) of the natives reported
overestimation of body image and 25,9% (CI 95% 21,4-31)
subestimation.In Afrodescendents 9,5% (CI 95% 7,4-12,1)
overestimate their BMI and 29% (CI 95% 25,5-32,8)subes-
timate it. In the adults group, the natives reported overesti-
mation in 3,8% (IC 95% 2-7,1) and 39% (IC 95% 33,8-44,5)
subestimation.Overestimation in native adults is significantly
lower than the afrodescendents and the rest of the population.
In contrast, subestimation is higher than in other groups.The
most prevalent risk behavior in natives was to use products to
lose weight and in the afrodescendent was to fast for 24 hours
or more.
Conclusion:
These findings suggest that some ethnic
minorities in Colombia seems to show similar risk factors as
the rest of the Colombian population , as have been reported
in other countries.
T58
CHARACTERISTICS OF WOMEN WITH BODY SIZE SAT-
ISFACTION AT MIDLIFE: RESULTS OF THE GENDER AND
BODY IMAGE STUDY (GABI)
Cristin D. Runfola
1
, Ann Von Holle
1
, Christine M. Peat
1
, Dan-
ielle A. Gagne
1
, Sara M. Hofmeier
1
, Cynthia M. Bulik
1,2
1
Department of Psychiatry, University of North Carolina at
Chapel Hill, Chapel Hill, NC, USA,
2
Department of Nutrition,
University of North Carolina at Chapel Hill, Chapel Hill, NC,
USA
Introduction:
Little is known about the intriguing minority of
women (~17%) who are satisfied with their body size. This
study characterizes the profile of women ages 50+ who report
body size satisfaction.
Method:
An online convenience sam-
ple of 1,849 women (mean age = 59.1, SD = 6.8) completed
a figure rating scale and answered questions about eating
disorders symptoms, weight and shape concerns, dieting and
weight control behaviors, and quality of life. Body size satis-
faction was defined as having a current body size equal to
ideal size. Permutation-adjusted
t-
tests and Fisher exact tests
were used to test differences between satisfied and dissatis-
fied women.
Results
: Satisfied women (12.2%; n = 218) were
less likely to be overweight/obese than dissatisfied women (7
vs. 63%). They were less likely to report trying to lose weight,
but exercised more. Weight and appearance dissatisfac-
tion, current eating disorders symptoms, and past unhealthy
dieting behaviors were less common in satisfied women.
Interestingly, weight and shape still played a primary role in
their self-evaluation, and weight monitoring and appearance
altering behaviors (e.g., Botox) did not differ between groups.
Conclusions:
Satisfied women report better overall function-
ing, but appear to exert considerable effort in achieving body
size satisfaction. Understanding the profile of satisfied women
may guide interventions aimed to improve body image and
enhance wellness in this growing, underserved population.
T59
IDENTIFYING THE CUT-OFF SCORE OF EATING DISOR-
DER EXAMINATION QUESTIONNAIRE TO PREDICT CASE
STATUS
Øyvind Rø, Deborah L. Reas
Regional Eating Disorder Service, Oslo University Hospital,
Oslo, Norway
Introduction:
The Eating Disorder Examination Question-
naire (EDE-Q) is a self-report version of the Eating Disorder
Examination interview used in epidemiologic and clinical
studies. A cut-off score of 4.0 on the global EDEQ score has
been used as a threshold of severe eating disorder (ED)
psychopathology. Normative EDE-Q data from clinical sam-
ples have shown that nearly half of patients with ED obtain
a global score of less than 4.0. The aim of this study is to
establish a cut-off score of the EDE-Q to determine caseness
and test the positive predictive values.
Methods:
A represen-
tative sample of 1058 community women aged 16-50 years
without ED and 577 patients seeking treatment at specialized
ED units in Norway were administered the EDE-Q (version
6.0). Receiver operating characteristic analyses were used to
determine a cut-off score with optimal sensitivity and specific-
ity.
Results:
A cut-off score of 2.5 had optimal sensitivity and
specificity of 86%. Assuming a prevalence of ED of 3% in the
community sample and a prevalence of 80% among patients
seeking treatment for ED at specialized units, results showed
a positive predictive value of 16% and 96%, respectively.
Conclusions:
The clinical cut-off score of 2.5 in this study
falls between the values proposed by Mond of 2.3 (Mond et
al. 2004) and 2.8 (Mond et al. 2008). Using this cut-off score
yields a very low positive predictive value in a community
sample, but an acceptably high value in a treatment-seeking
population.
T60
Anorexia Nervosa: what never recovers
Nicole E.P. Sanders
1,2
, Paul A.M. Smeets
3
, Unna N. Dan-
ner
1,4
, Annemarie A. van Elburg
1,5
, Hans W Hoek
6
, Roger A.H.
Adan
1,2
1
Rintveld Center Eating Disorders, Altrecht Mental Health
Institute, Zeist, Netherlands,
2
Rudolf Magnus Institute of
Medical Science, University Medical Center Utrecht, Utrecht,
Netherlands,
3
Image Sciences Institute , University Medical
Center Utrecht, Utrecht, Netherlands,
4
Clinical and Health
Psychology, Utrecht University, Utrecht, Netherlands,
5
Dept
of Child & Adolescent Psychiatry, University Medical Center
Utrecht, Utrecht, Netherlands,
6
Parnassia Bavo Psychiatric
Institute, The Hague, Netherlands
Introduction:
Anorexia nervosa (AN) is characterized by
obsessive behavior with regard to food and body weight.
Unlike healthy subjects, women with AN are able to starve
themselves by actively restricting their food intake. Differenc-
es in introceptive awareness, hormone concentrations and
brain activation patterns have been suggested to explain this
behavior. Here we aim to compare eating behavior in chron-
ically ill and recovered women with AN and healthy controls.
Methods:
Three groups (n=15 each) participated: women
with chronic AN (c-AN), women recovered from AN (r-AN)
and healthy controls (HC). A blood sample was drawn after
an overnight fast. Subjects underwent an fMRI scan at 3T,
during which images of foods and non-foods were shown.
POSTER SESSION 1 ABSTRACTS