Basic HTML Version
EDRS 2012
28
sures, zinc deficiency, environmental temperature, maturity of
the amygdala. However, as described by Kenneth Rothman
in the 1970s, most diseases are eventually found to have
several separate causal pathways, with both common and
different components. Diseases from cancer to amyloidosis
have become subdivided by cause. Once different groups
of causal factors are known; prevention, diagnosis, and
treatment can be individualized and, for some, much more
treatable. Decision analysis computer programs, like D-Maker
using Markov analysis, can explicitly separate factors, assign
probabilities that can be changed throughout life, and when
combined with the populations demographys, can predict
the incidence and prevalence in subpopulations. Comparing
these predictions to known data will allow an iterative deci-
sion making process about the relevance of risk factors. For
example, about 33% of women by the age of 24 have a his-
tory of sexually abuse (Golder and Birmingham). Obviously,
33% of women don’t get AN. The timing of hormonal chang-
es, changes in diet, etc can be introduced that a model can
gradually be developed and then prospectively tested. This
is the first project of its kind in the world. 2. Methods. obtain
information on possible causal factors of anorexia nervosa
(age, sex, percent),obtain population demographics for Can-
ada, and construct a Markovdecision analysis model. Initially
all individual factors will be entered, then combinations for
which there is evidence or logic, and finally groupings accord-
ing to the model of Kenneth Rothman. 3. This study is just
beginning. At the time of the meeting I will present Markov
modelling, and data we will use regarding risk factors, and at
least some samples of incidence and prevalence predictions.
T10
OVEREATING PHENOTYPES IN OVERWEIGHT AND
OBESE CHILDREN
Kerri N. Boutelle
1
, Guy Cafri
1
, Nancy Zucker
2
, Carol B. Peter-
son
3
, Sarah A. Rydell
3
, Lisa Harnack
3
1
University of California, San Diego, CA, USA,
2
Duke Univer-
sity, Durham, NC, USA,
3
University of Minnesota, Minneapo-
lis, MN, USA
The purpose of this study was to identify overeating pheno-
types and their correlates in overweight children. Sample
included 118 overweight and obese 8-12 year old children
and their parents.Children completed all of the following
measures: Eating in the absence of hunger (EAH) paradigm,
EDE interview, 3 24-hour dietary recalls, self-report of satiety
responsiveness, enjoyment of food, EAH, and height and
weight. Parents reported on child’s EAH and completed 5
subscales of the Child Eating Behaviour Questionnaire: Food
Responsiveness, Emotional Overeating, Enjoyment of Food,
Satiety Responsiveness, and Slowness in Eating. Latent
class analysis was used to identify heterogeneity in overeat-
ing phenotypes in the child participants and were compared
on demographics, obesity status and nutritional intake.Three
latent classes of overweight and obese children were iden-
tified: Cognitive & Emotional Overeating (CE), Food Cue Sen-
sitive & Reward Overeating (FR), and General Overeating
(GO). Results showed that the FR class had higher BMI and
BMI-Z scores compared to the GO class. No differences were
found among classes in demographics or nutritional intake.
This study identified 3 overeating phenotypes supporting the
heterogeneity of eating patterns associated with overweight
and obesity in children. These phenotypes can potentially
be used to identify high risk groups, inform prevention and
intervention targets, and identify specific treatments for these
phenotypes.
T11
FOOD CHOICE AND DISORDERS OF EATING BEHAVIOR:
CORRELATIONS WITH THE PSYCHOPATHOLOGICAL
ASPECTS OF THE DISEASES
Francesca Brambilla
1
, Cristina Segura-Garcia
2
, Flora Sinop-
oli
2
1
Psychoneuroendocrinology Center, Milano, Italy,
2
Dept.
Health Sciences, University Magna Graecia of Catanzaro,
Catanzaro, Italy
Purpose of the study. Food choice of patients with Anorex-
ia Nervosa (AN), Bulimia Nervosa (BN) and Binge-Eating
Disorder (BED), its macronutrients composition, and cor-
relation with psychopathology of the diseases was studied,
as suggestion for the biopathological background of Eating
Disorders. Methods. 37 AN-restricter 18 AN-bingeing/purging,
40 BN and 29 BED female patients and 20 controls filled a 7
days dietary diary and were administered EDI-2 and TCI-R
scales to examine the psychological aspects of the diseases.
Multiple Regression Analysis was performed to assess the
association between macronutrient choices and psychological
characters. Results.
C
ompared to controls, all the patients
showed
significantly lower intake of animal proteins lower lac-
toproteins in BN and BED, higher vegetable proteins, lower
fats and normal oligosaccharides percent in AN-R. Significant
positive correlation were observed between proteins, fat and
carbohydrates consumption and Novelty Seeking and Coop-
erativeness and negative with Drive for Thinness. Complex
carbohydrates correlated positively with Harm Avoidance and
negatively with Bulimia, Social Insecurity and Perfectionism,
Conclusions
.
food choice in Eating Disorders might depend
from the alterations of neurotransmitters-neuropeptides-pe-
ripheral peptides which both regulate and are regulated by
macronutrients intake and which are also the basis of the
psychological and temperamental alterations
T12
PTSD AND PARTIAL PTSD AS MEDIATORS BETWEEN
EATING DISORDERS AND SUBSTANCE USE DISORDERS
AMONG WOMEN AND MEN IN THE NATIONAL COMOR-
BIDITY SURVEY-REPLICATION STUDY
Timothy D Brewerton
1
, Karen S Mitchell
2
1
Department of Psychiatry & Behavioral Sciences, Medical
University of South Carolina, Charleston, SC, USA,
2
Wom-
en’s Health Sciences Division, National Center for PTSD &
Department of Psychiatry, Boston University, Boston, MA,
USA
Introduction: Individuals with EDs characterized by binge-
ing &/or purging have been reported to have higher rates of
substance use disorders (SUDs), and conversely, individuals
with SUDs have higher rates of EDs. Each group of disorders
has been reported to have high rates of trauma and resulting
PTSD or partial PTSD (pPTSD). PTSD has been reported
to be a powerful mediator between EDs and SUDs in other
studies. Methods: Using data generated from the National
Comorbidity Survey-Replication Study, we tested the hypoth-
esis that PTSD and pPTSD act as mediators of associations
between the various EDs and SUDs (alcohol abuse, alcohol
dependence, drug abuse, drug dependence, and a composite
variable representing endorsement of “any SUD”). Results: In
women, PTSD and pPTSD were significant mediators of BN
and alcohol dependence, drug abuse, drug dependence, and
any SUD. PTSD and pPTSD were significant mediators of
any bingeing and drug dependence, and any SUD. PTSD and
pPTSD did not mediate the relations between BED or AN and
any of the SUDs. In men, PTSD and pPTSD were significant
mediators of BED and alcohol dependence, drug abuse, drug
dependence, and any SUD. PTSD and pPTSD were also sig-
nificant mediators of any bingeing and any SUD. Conclusion:
These results confirm the considerable overlap between EDs
(with bingeing &/or purging) and most SUDs. Implications for
neurobiological mechanisms and treatment approaches in
light of new ASAM definitions of addictions will be discussed.
POSTER SESSION 1 ABSTRACTS