Basic HTML Version
Porto, Portugal | September 20-22, 2012
15
1
EATING DISORDERS OVER TIME AND PLACE AND THE
FUTURE OF EPIDEMIOLOGICAL RESEARCH
Hans W. Hoek
1,2,3
1
Parnassia Bavo Psychiatric Institute, The Hague,
Netherlands,
2
University Medical Center Groningen,
Groningen, Netherlands,
3
Columbia University, New York, NY,
USA
This lecture will review our epidemiological studies and
how our research group tries to understand the nature and
causes of eating disorders. In the last century the validity of
epidemiological studies of eating disorders has sometimes
been called into question by methodological problems
relating to the selection of populations under study and the
identification of cases. In the present era community based
studies and improved diagnostic reliability along with case-
control studies or even cohort studies allow a shift to risk
factor epidemiology. Many of our eating disorder studies
focused, like most epidemiological studies, on prevalence
and incidence rates as the basic measures of disease
frequency*. Our new data show that the incidence of anorexia
nervosa remained stable over the past four decades, while
the incidence of bulimia nervosa decreased significantly
over the past three decades. Eating disorders are likely the
result of multilevel causation; the biological underpinnings
of anorexia nervosa are strong, but the societal roots are
also compelling**. In several cohort studies we are currently
searching for the causes of eating disorders. We hope
these studies will contribute to understanding and ultimately
controlling the risk factors of eating disorders. *
Hoek et al
J Psychiatr Res 1985, Psychol Med 1991, Am J Psychiatry
1995, World Psychiatry 2009; Van Son et al Br J Psych
2006; Keski et al Am J Psychiatry 2007, Psychol Med 2009;
Machado et al Int J Eat Disord 2009, 2012; Raevuori et al
PLoS ONE 2009; Van Hoeken et al Int J Eat Disord 2009, Eur
Eat Disord Rev 2010; Smink et al Curr Psychiatry Rep 2012.
**
Hoek et al N Eng J Med 1998; Hoek et al Am J Psychiatry
2005
2
DSM-5 eating disorders in a ten year cohort
study of adolescents
Frederique Smink
1
, Daphne van Hoeken
1
, Iris H. van der
Meer
1
, Albertine J. Oldehinkel
2
, Hans W. Hoek
1,2,3
1
Parnassia Bavo Psychiatric Institute, The Hague,
Netherlands,
2
Department of Psychiatry, University Medical
Center Groningen, Groningen, Netherlands,
3
Department of
Epidemiology, Columbia University, Mailman School of Public
Health, New York, NY, USA
Introduction The epidemiology of DSM-IV eating disorders,
especially AN and BN, has been extensively studied
1
. Less
is known about the occurrence of eating disorders according
to the proposed DSM-5-criteria. Our aim is to study the
prevalence, course and risk factors of DSM-5 eating disorders
in a community cohort of adolescents. Method This study
is part of TRAILS (TRacking Adolescents’ Individual Lives
Survey), a longitudinal cohort study on mental health and
social development from preadolescence into adulthood. The
participants (N=2230) enrolled in the study at age 10-11. At
the 4th assessment wave (age 19) the CIDI was completed
in 1584 subjects. BMI was assessed at all 4 waves. Singular
strengths of the study are prepubertal baseline measures,
high response rates and a very extensive database, covering
a wide biopsychosocial spectrum. A group at high risk for
eating disorders was selected for an interview by eating
disorder experts at age 20, aimed at diagnosing DSM-5
eating disorders. Results The high-risk group comprises 312
subjects, of which 296 could be interviewed (response rate
95%). The most common diagnoses are AN (16 cases) and
BED (25). Other diagnoses found are BN (6) and Feeding
Disorder (3 ARFID and 1 Pica). A minority of cases fall into
the rest category of FECNEC (2 subthreshold ED, 1 NES and
1 atypical ED; 7.3% of all ED diagnoses). Among 20-year-old
females the lifetime prevalence of AN is 1.8% and of BED
2.5%. None of the AN cases were detected by the CIDI; of
the 31 BN cases diagnosed by the CIDI, 28 cases could be
interviewed, of which only 2 had a true lifetime diagnosis
of BN. Conclusion The proposed DSM-5-criteria of eating
disorders appear to be successful in reducing the size of the
EDNOS-category. The CIDI has low sensitivity and specificity
for diagnosing eating disorders.
3
Eating disorders in the “noughties”: a UK
incidence study in primary care
Nadia Micali
Behavioural and Brain Sciences Unit, Institute of Child health,
UCL, London, United Kingdom
Background: Investigating time-trends in the incidence of
eating disorders (ED) can help our understanding of disease
patterns; important questions about changes in incidence
over time and case detection remain unanswered. Aims: To
determine changes in the incidence of ED: anorexia nervosa
(AN), bulimia nervosa (BN) and eating disorder not otherwise
specified (EDNOS) between 2000-2009 in the UK and to
describe referral patterns to specialist care. Method: We
identified all patients with a first time diagnosis of AN, BN
and EDNOS from the General Practice Research Database.
Annual Incidence rates (IR) were calculated by age group,
gender, and ED sub-type. Results: Annual IR of all ED for
ages 10-49 ranged between 31.4-37.1 per 100,000 during
the study period. In females the incidence of AN and BN was
stable, however the incidence of EDNOS increased. Between
2000 and 2009 there was a non-significant increase in the
ED annual IR in males. Seventy percent of AN cases were
referred to secondary and tertiary care, compared to 30%
of BN and EDNOS. Conclusions: The incidence of AN and
BN remained stable between 2000 and 2009. There was
a significant increase in new diagnoses of EDNOS among
females. EDNOS is the most common ED in primary care.
4
DSM-5 reduces EDNOS: Evidence from
community samples
Paulo Machado
1
, Sónia Gonçalves
1
, Hans W. Hoek
2,3,4
1
Psychotherapy and Psychopathology Research Unit – CIPsi,
School of Psychology, University of Minho, Braga, Portugal,
2
Parnassia Psychiatric Institute, The Hague, Netherlands,
3
Department of Epidemiology, Columbia University, Mailman
School of Public Health, New York, NY, USA,
4
Department of
Psychiatry, University of Groningen, Groningen, Netherlands
Objective: Eating Disorder Not Otherwise Specified (EDNOS)
constitute the most common eating disorder among those
seeking treatment at eating disorder facilities (40-60%); they
are even more common among persons with eating disorders
in non-specialty settings and in the community. This study
compares the impact of applying the revised diagnostic
criteria proposed by the DSM-5 workgroup, and the broad
categories for the diagnosis of eating disorders (BCD-ED)
proposed by Walsh and Sysko on the prevalence of EDNOS
cases. Method: In two nationwide epidemiological studies in
Portugal the prevalence of eating disorders among female
high school (n=2028) and university students (n=1020)
was examined using DSM-IV criteria. We used a two stage
design, administering a questionnaire (EDE-Q) in the first
stage and an interview (EDE) in the second stage. Results:
In the combined samples 118 cases of eating disorders
(DSM-IV) were detected, of which 86 were diagnosed as
EDNOS (72.9%). Application of the DSM-5 criteria reduced
the number of EDNOS cases to 60 (50.8%); with the use of
BCD-ED criteria, only 5 (4.2%) cases of EDNOS remained.
Discussion: Proposed criteria set for DSM-5 substantially
reduce the number of EDNOS cases. However, the BCD-ED
scheme further reduces its proportion, almost eliminating it.
PLENARY/ORAL ABSTRACTS
PLENARY/ORAL ABSTRACTS