Content uploaded by David Kolar
Author content
All content in this area was uploaded by David Kolar on Oct 05, 2017
Content may be subject to copyright.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
C
URRENT
O
PINION
Epidemiology of eating disorders in Latin America:
a systematic review and meta-analysis
David R. Kolar
a,b
, Dania L. Mejı´a Rodriguez
b,c
, Moises Mebarak Chams
b
,
and Hans W. Hoek
d,e,f
Purpose of review
Eating disorders are currently not considered to be limited to Western culture. We systematically reviewed
the existing literature on the prevalence of eating disorders in Latin America.
Recent findings
Of 1583 records screened, 17 studies from Argentina, Brazil, Chile, Colombia, Mexico, and Venezuela
were included in the analysis. Most studies reported point-prevalence rates and only three studies provided
lifetime and 12-month prevalence rates. We found a mean point-prevalence rate of 0.1% for anorexia
nervosa, 1.16% for bulimia nervosa, and 3.53% for binge-eating disorder (BED) in the general population.
Heterogeneity for bulimia nervosa and BED was large. This meta-analysis indicates that the prevalence of
anorexia nervosa seems to be lower, whereas the prevalence of bulimia nervosa and especially of BED
seems to be higher in Latin America than in Western countries.
Summary
Our findings show that eating disorders are common mental disorders in Latin America. However, some
facets of Latin American culture might be protective for the development of anorexia nervosa and increase
the risk for bulimia nervosa and BED. Further studies investigating the epidemiology of eating disorders and
their relation to culture in Latin America are needed.
Video abstract
http://links.lww.com/YCO/A35
Spanish abstract
http://links.lww.com/YCO/A36
Keywords
anorexia nervosa, binge-eating disorder, bulimia nervosa, epidemiology, Latin America
INTRODUCTION
Historically, eating disorders were perceived as
culture-bound syndromes [1] restricted to the West-
ern culture. Eating disorders were first described in
Caucasian females living in Western Europe and
North America, leading to the presumption that
specific characteristics of their culture must be cru-
cial to the development of an eating disorder. Sev-
eral decades after the emergence of eating disorders
in Western culture, cases of eating disorders have
been identified in all cultures to a varying extent,
mostly with increasing incidence rates but still lower
prevalence rates than in Western countries [2,3].
Recent evidence suggests that the incidence of
bulimia nervosa is decreasing in Western countries,
but remains stable for anorexia nervosa [4,5]. The
point-prevalence for anorexia nervosa in Western
Europe and North America ranges from 0 to 0.9%
a
Department for Child and Adolescent Psychiatry and Psychotherapy,
University Medical Center of the Johannes Gutenberg University, Mainz,
Germany,
b
Departamento de Psicologı´a, Universidad del Norte, Barran-
quilla, Colombia,
c
Departamento de Ciencias Sociales y Humanas,
Universidad Simo
´n Bolı´var, Barranquilla, Colombia,
d
Department of
Epidemiology, Mailman School of Public Health, Columbia University,
New York, New York, USA,
e
Parnassia Psychiatric Institute, The Hague,
The Netherlands and
f
Department of Psychiatry, University Medical
Center Groningen, University of Groningen, Groningen,The Netherlands
Correspondence to David R. Kolar, Department of Child and Adolescent
Psychiatry and Psychotherapy, University Medical Center, Langen-
beckstr., 1 55131 Mainz, Germany. Tel: +49 6131 17 3280; fax: +49
6131 17 5580; e-mail: david.kolar@unimedizin-mainz.de
Curr Opin Psychiatry 2016, 29:363– 371
DOI:10.1097/YCO.0000000000000279
This is an open-access article distributed under the terms of the Creative
Commons Attribution-NonCommercial-NoDerivatives 4.0 License, where
it is permissible to download and share the work provided it is properly
cited. The work cannot be changed in any way or used commercially.
0951-7367 Copyright ß2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com
REVIEW
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
with a mean estimate of 0.29% in the high-risk
population of young females [6,7]. For bulimia
nervosa, a general point-prevalence of up to 1%
among young females is widely recognized [7–9].
Regarding eating disorders in Asia, the prevalence
rates in Japan and China are now considered at par
with European levels of eating disorders [10,11],
whereas eating disorders seem to be at a rise in Arab
countries, but still have lower prevalence rates com-
pared with Western nations [12]. The proceeding
industrialization, globalization and therefore west-
ernization in these regions are suspected as catalysts
for increasing incidence rates of eating disorders
[2,13
&
].Inconsiderationofthecloseproximityof
Latin America to the United States, not only in a
geographical but also economical manner, research
on the epidemiology of eating disorders in these
countries could further confirm the westernization
theory of rising prevalence rates of eating disorders
in developing countries. Recent research suggests
that prevalence rates of eating disorders in urban-
ized regions of Latin America reach similar levels as
in Europe and the United States [14,15], and Latin
American immigrants in Western countries did
not differ from the Caucasian population regarding
diagnoses of eating disorders [16,17]. However,
English language literature on the epidemiology of
eating disorders in Latin America is still rare, as most
studies are published in Spanish or Portuguese.
As a first step to better understanding eating
disorders in Latin America, this study provides
a systematic review of the existing literature
on the epidemiology of eating disorders in the
general population in Latin America. Its focus
is on the prevalence of anorexia nervosa, bulimia
nervosa and binge-eating disorder (BED) in Latin
American countries in continental South and
Central America.
METHODS
The review was conducted in accordance to the
Preferred Reporting Items for Systematic reviews
and Meta-Analyses statement (PRISMA) [18].
This review focuses on the prevalence rates of ano-
rexia nervosa, bulimia nervosa, and BED in Latin
American countries. Latin America is defined as all
American countries in which Romance languages
are spoken. For a more focused review, only
continental American countries were considered.
Therefore, literature on the prevalence of eating
disorders in the following countries was included:
Argentina, Bolivia, Brazil, Chile, Colombia, Costa
Rica, Ecuador, El Salvador, Guatemala, Honduras,
Mexico, Nicaragua, Panama, Paraguay, Peru, Uru-
guay, and Venezuela. Suriname, Guyana, and Belize
were excluded, as their official languages are not
considered Romance. French Guyana was excluded,
as it is not a sovereign state.
Literature search
A systematic literature search was conducted in
January 2015 and was updated in May 2016 for
recently published articles. The following databases
were screened: Medline (via PubMed and DIMDI),
Web of Science, SciELO, LILACS, IBECS, INDPSI,
HISA, and LIS. The search term was a combination
of the disorders of interest (eating disorder, anorexia
nervosa, bulimia nervosa, and BED), ‘South Amer-
ica’ and ‘Latin America’, and all names of the
countries included. Because of the interest in local
literature, SciELO, LILACS, IBECS, INDPSI, HISA,
and LIS were searched in Spanish and Portuguese
in addition to English.
Data extraction and quality assessment
Records identified in the literature search were
evaluated in a two-step approach. A screening of
titles and abstracts of all publications found in the
databases was conducted by two researchers. Articles
fulfilling all of the following eligibility criteria were
considered for full-text review: (1) the article was
written in English, Spanish, or Portuguese; (2) the
article was related to eating disorders; and (3)
the article was related to the epidemiology of
these disorders. Articles considered as relevant were
reviewed in full-text. Articles meeting at least one
exclusion criterion of the following were excluded
during screening and full-text review: (1) editorials,
newspaper articles, literature reviews, study pro-
tocols, theory articles, unpublished manuscripts,
KEY POINTS
Few cases of anorexia nervosa were found in the
epidemiological studies analyzed in this review,
indicating that the Latin American culture might be a
protective factor.
The prevalence rate of bulimia nervosa might be
slightly underestimated in our review and is likely to be
higher than in Western Europe or North America.
Binge-eating disorder appears to have a high
prevalence in Latin America.
The English literature on eating disorders in Latin
America is scarce as most of the studies reviewed were
published in either Spanish or Portuguese language
journals, which decreases the international visibility of
studies on eating disorders in Latin America.
Eating disorders
364 www.co-psychiatry.com Volume 29 Number 6 November 2016
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
dissertations, government reports, books and book
chapters, conference proceedings, meeting
abstracts, and lectures; (2) the article was published
in a journal without peer review; (3) samples with
overall sample age under 10; (4) studies with biased
sample selection; or (5) the article studied a clinical
population. Articles fulfilling the inclusion criteria
as provided in Table 1 were included in a
qualitative analysis.
The screening and full-text review was con-
ducted online using Covidence (Covidence.org,
Alfred Health, Melbourne, Australia). Two research-
ers decided on the inclusion of each article at each
stage of the review. If disagreement occurred, resolv-
ing was conducted by the principal investigator. The
first three authors speak fluent English and Spanish,
and the first author speaks Portuguese as well. In
the case of difficulties understanding Portuguese
language articles, a translation was given by the first
author.
Quality assessment of the articles was conducted
with the Strengthening the Reporting of Observa-
tional Studies in Epidemiology (STROBE) checklist
[19]. Additionally, a methodological scoring system
to rate studies adapted from Loney et al. [20]
was used.
Statistical analyses
As three disorders were analyzed in this review,
separate meta-analyses were conducted for each
disorder. Only two studies reported 12-month
and lifetime prevalence rates, hence quantitative
analyses were conducted for point-prevalence rates
only. In many epidemiological studies, prevalence
rates are assessed by first screening a larger popu-
lation with a screening questionnaire to identify
people at risk for having an eating disorder. Sub-
sequently, the final diagnosis of an eating disorder
is established in a personal interview to which only
the high-risk group is invited. The prevalence rates
of these so-called two-stage studies included in the
meta-analysis were recalculated and based on
participants screened at first stage, as not all studies
provided response rates for the second stage. If,
for example, 2770 patients were included, 2756
participants filled in the screening instrument and
23 cases were detected at the second stage, the preva-
lence rate would be p¼23/2756. Meta-analyses were
conducted with MetaXL 3.0 (epigear.com), which
allows analyzing a quality effects model accounting
for heterogeneous prevalence studies, by taking the
overall quality scores of the studies into account [21].
This model was calculated for each disorder separ-
ately. As recommended by Barendregt et al. [22],
prevalence values were normalized and double arc-
sine transformed to stabilize variance of prevalence
estimates near 0. Finally, forest plots presenting the
overall and study-specific prevalence, 95% confi-
dence intervals and the study weight were computed.
RESULTS
In our literature search, 2199 articles were ident-
ified as shown in the study flow diagram (Fig. 1).
After removal of duplicates, 1583 articles were
Table 1. Inclusion criteria for articles
Inclusion criterion Definition
1 Type of eating disorder
investigated
Only anorexia nervosa, bulimia
nervosa and/or BED
2 Origin of study sample Sample recruited in one of the previous
defined Latin American countries
a
3 Sample size More than 200 subjects screened
3 Type of diagnosis
assessment
Only questionnaires and/or interviews
based on DSM-IV, DSM-V or ICD-10
criteria for anorexia nervosa, bulimia
nervosa and/or BED
4 Reporting of
prevalence
Prevalence rates for at least one of the
diagnoses of interest must be reported
BED, binge-eating disorder.
a
Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, El
Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay,
Peru, Uruguay, and Venezuela.
Medline
via PubMed: 791
via DIMDI: 100
ISI web of
knowledge
484
SciELO
328
LILACS
496
2199 records
identified through
database searching
129 records
identified
through other
sources
1583 records
screened after
duplicates removed
1361 records
excluded
222 full-text articles
assessed for
eligibility
205 records excluded:
98 only subclinical
disordered eating
studied
28 clinical population
18 missing prevalence
13 sample age or size
13 sample selection biased
10 no valid instrument
9 other disorders studied
5 no Latin American
sample
4 no peer-review
3 same population
3 study protocol
1 Editorial
17 studies included in
qualitative synthesis
15 studies included in
quantitative syntheses
(meta-analyses)
FIGURE 1. Study flow diagram. IBECS, INDPSI, HISA, and
LIS were searched in combination with LILACS.
Epidemiology of eating disorders in Latin America Kolar et al.
0951-7367 Copyright ß2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com 365
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Table 2. Sample characteristics, measures, prevalences and quality scores of all studies included
Study and
setting
Sampling
frame
Sample
design
a
(age)
Sample
size (n) Measures
b
Response
rate
Prevalence rates Score
Anorexia
nervosa
c
Bulimia
nervosa
c
BED
c
Angel 2008 [23]
Colombia – urban
Municipal educational
census
Whole population of high school
students (mean age 14.2 years)
2770 2-phase design: 1.
ECA; 2. SCAN
33.5% response
in phase 2
0.11% 0.83% n/a 4
Baader 2014 [24]
Chile – Valdivia
Census of the Universidad
Auto
´noma de Chile;
stratified by subjects and
classes
University students (mean age 21
years)
804 EDDS 97.4% 0.13% 6.13% 2.55% 6
Benjet 2012 [25,26]
Mexico – rural/
urban
National Census; stratified
for seventeen districts
of Mexico
Adolescents (12– 17 years) 3005 WMH-CIDI 71% 0.5%
d
;
0.1%
e
1%
d
; 0.7%
e
1.4%
d
; 0.7%
e
8
Compte 2015 [27
&&
]
Argentina – urban
Randomized selection of 12
universities in Buenos Aires;
six universities participated
All male students (18– 28 years) of
participating classes
475 2-phase design: 1.
EAT-26; 2. EDE interview
99.4% in first
phase, 77.8%
in second
phase
0% 0% 0% 7
Freitas 2008 [28]
Brazil – Rio de
Janeiro
Census (Brazilian Institute of
Geography and Statistics)
Female individuals 35 years 1500 SCID-I/P questions 86.5% n/a n/a 9.78% 6
Herscovici 2005
[29,30] Argentina –
rural and urban
Pediatric primary care
distributed accordingly to
national census
Children and adolescents (10– 19
years)
1971 2-Phase design: 1.
EDE-Q4; 2. EDE
interview
42% response
rate in
second phase
0% 0.05% 6.60% 5
Kessler 2013 [14]
Colombia – urban
community
WHO World Mental Health
(WMH) Survey Initiative;
sample representative for
urbanized areas
(Colombia/Mexico) or one
specific urbanized area
(Sa
˜o Paulo, Brazil)
Adults (18 years) 1217 WMH-CIDI (DSM-V
criteria for BED)
Response rates
range from
45.9 to
87.7% and
average at
68.8%.
n/a 0.4%
d
; 0.2%
e
0.9%
d
; 0.3%
e
5
Brazil – Sa
˜o Paulo
community
2942 n/a 2%
d
; 0.9%
e
4.7%
d
; 1.8%
e
Mexico – urban
community [31]
1236 n/a 0.8%
d
; 0.3%
e
1.6%
d
; 0.5%
e
Mancilla-Diaz 2007
[32] Mexico –
Mexico city
university students
Probabilistic and stratified
sample based on all female
university students in 2002;
not further explained
Female university students 1995
sample: (mean age 19.49 years)
522 2-phase design:
1.EAT-40, BULIT
screening; 2.
IDED – IV
32.9% 0 1.15% n/a 3
2002 sample: (mean age 19.53
years)
880 40.3% 0 1.14% n/a
Mascarenhas 2007
[33,34] Brazil –
Feira da Santana,
BA
Randomized sample, stratified
for subdistricts of Feira da
Santana
Adults (20– 59 years) 2506 PHQ (2 binges per
week for 3 months
criterion)
n/a n/a 1.00% 5.03% 4
Morales 2015 [35
&
]
Venezuela –
urban
Census of all enrolled medical
students at the local
university (UCLA)
Randomized sample of medicine
students (18– 28 years)
497 Questionnaire
based on DSM-V
criteria for BED
n/a 3.20% 5
Pin
˜eros-Ortiz
2010 [36]
Colombia –
urban community
Randomized selection of
education institutes in the
region.
Adolescents (12– 20 years) 937 2-phase design: 1.
EAT-26; 2.
telephone
interview
85.18% in first
phase; 38.3%
in second
phase
0% 0.21% n/a 3
Pivetta 2010 [37]
Brazil – high school
students
Randomized selection of
registered education
institutes, stratified for
public and private.
Adolescents (14– 19 years) 1209 QEWP-R 93.3% n/a 1.90% 1.82% 4
Quintero-Pa
´rraga
2003 [38]
Venezuela –
Maracaibo city
Educational population
reported. Official State
Census
Middle and high school Students (12–
18 years)
1363 The A.B.A checklist 80.86% 0% 1.54% 0.66%; 4
Rueda-Jaimes 2005a
[39] Colombia –
Bucaramanga
Three schools selected at
convenience from different
socio-economic districts of
Bucaramanga
All female school students (10– 19
years)
247 WMH-CIDI 98.4% 1.65% 5.35% n/a 6
Rueda-Jaimes 2005b
[40] Colombia –
Bucaramanga
Six faculties with
predominantly female
students of the Universidad
Auto
´noma de Bucaramanga
Female students in 2nd semester 241 2-phase design:
1. SCOFF; 2.
WMH-CIDI
62.6% 0% 2.08% n/a 5
Rueda-Jaimes 2008
[41] Colombia –
Bucaramanga
Private Bucaramanga
University
University students (16– 35 years) 261 WMH-CIDI 99.61% n/a 1.15% 4.21% 6
Va
´zquez 2005 [42]
Mexico – Mexico
City
Six schools in Mexico City School students (mean age 15.25
years)
525 2-phase design:
1. EAT-40;
Bulit; 2. IDED
n/a 0.38% 0.95% n/a 3
BED, binge-eating disorder; BULIT, bulimia Test; EAT, Eating Attitudes Test; ECA, Encuesta de Comportamento Alimenticio; EDDS, Eating Disorder Diagnostic
Scale; EDE, Eating Disorder Examination; IDED, Interview for Diagnosis of Eating Disorders; PHQ, Patient Health Questionnaire; QEWP-R, Questiona
´rio sobre
Padro
˜es de Alimentac¸a
˜o e Peso – Revisado; SCAN, Cuestionarios para la Evaluacio
´nClı
´nica en Neuropsiquiatrı
´a; WMH-CIDI, World Mental Health Composite
International Diagnostic Interview.
a
Random sample, if not specified.
b
Measurements based on DSM-IV, if not specified.
c
Prevalences are reported as point prevalence, if not specified.
d
Lifetime prevalence.
e
12-month prevalence.
Eating disorders
366 www.co-psychiatry.com Volume 29 Number 6 November 2016
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
screened and 1361 articles removed as irrelevant.
A total of 222 studies were reviewed in full-text,
resulting in an inclusion of 17 studies. Reasons for
the exclusion of studies are given in Fig. 1. Fifteen
of the studies provided point-prevalence rates of
eating disorders and were therefore included in the
meta-analyses.
Table 2 provides sample characteristics,
measures and reported prevalence values of each
study included. Of the studies included, five
were conducted in Colombia, three in Brazil, three
in Mexico, two in Venezuela and one each in
Argentina and Chile. One additional study pro-
vided data from Colombia, Brazil, and Mexico.
Due to the small number of studies, an inter-
national comparison of the results was not
conducted.
Lifetime and 12-month prevalence rates
Only two of the included studies reported lifetime
and 12-month prevalence rates for bulimia nervosa
and BED [14,25], and only one study for anorexia
nervosa [25]. Regarding Mexico, lifetime prevalence
rates for adolescents aged 12–17 years were 0.5%
(12 months: 0.1%) for anorexia nervosa, 1.0%
(0.7%) for bulimia nervosa, and 1.4% (0.7%) for
BED [25]. In a study on adults, lifetime prevalence
rates of bulimia nervosa were identified as 0.8% (12
months: 0.3%) in Mexico, 0.4% (0.2%) in Colom-
bia, and 2.0% (0.9%) in Brazil [14]. Regarding BED,
the same study identified the lifetime prevalence
ratesas1.6%(12months:0.5%)forMexico,0.9%
(0.3%) for Colombia, and 4.7% (1.8%) for Brazil
[14]. Lifetime prevalence rates were highest for
bulimia nervosa and BED in Brazil, and lowest in
Colombia [14].
Mean point-prevalence and heterogeneity
testing for anorexia nervosa
In 10 studies providing 11 distinct samples
(N¼10 840) assessing anorexia nervosa, there was
a weighed mean prevalence rate of 0.1%, 95% CI
(0.02, 0.23) for the general population in Latin
America (Fig. 2). The included studies varied
regarding their quality index from three to seven
points. Heterogeneity of the studies was moderate
(I
2
¼51.50). By exclusion of the most extreme
outlier Rueda-Jaimes et al. [39], the heterogeneity
can be reduced to almost 0. There was no real
difference between the mean prevalence in the
general population and that of a meta-analysis with
the female subsample only (0.1%, 95% CI [0.02,
0.35]; N¼6 334). Prevalence rates were comparable
amongst the different countries included.
Mean point-prevalence and heterogeneity
testing for bulimia nervosa
Regarding bulimia nervosa, 14 distinct samples in
13 studies were identified, with an overall sample
size of N¼14 816 participants assessed at first stage.
In the meta-analysis, a weighed mean prevalence
rate of 1.16% (95% CI [0.55, 1.98]) was calculated for
Latin America (Fig. 3). Heterogeneity was large
(I
2
¼92.28), reflecting the different sample settings
and countries included. The highest prevalence rate
was found for Chile, and the lowest for Argentina,
especially in a male sample [27
&&
]. The prevalence
rates for Colombia, Brazil, Venezuela, and Mexico
seem comparable. As the Chilean study was
conducted as a self-report screening, the high preva-
lence rate might be due to the study design. How-
ever, no study could be determined as a main source
of heterogeneity. In consideration of the small
Study Prev LCL UCL Weight %
Angel 2008
Baader 2014
Compte 2015
Herscovici 2005
Mancilla-Diaz 2007a
Mancilla-Diaz 2007b
Piñeros-Ortiz 2010
Quintero-Párraga 2003
Rueda-Jaimes 2005a
Rueda-Jaimes 2005b
Vázquez 2005
Overall
0.11
0.13
0.00
0.00
0.00
0.00
0.00
0.00
1.65
0.00
0.38
0.10 0.02 0.23 100.00
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.35
0.00
0.01
0.28
0.55
0.36
0.09
0.33
0.20
0.18
0.13
3.71
0.45
1.15
17.71
10.96
9.92
16.96
4.45
5.87
6.09
10.37
6.69
6.51
4.46
0.0 0.5 1.0
Prevalence
1.5 2.0
FIGURE 2. Forest plot of the meta-analysis of the anorexia nervosa point-prevalence rates. LCL, lower confidence level; Prev:
prevalence rate; UCL, upper confidence level.
Epidemiology of eating disorders in Latin America Kolar et al.
0951-7367 Copyright ß2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com 367
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
number of studies included, no moderator analysis
for heterogeneity was conducted.
Mean point-prevalence and heterogeneity
testing for binge-eating disorder
Within nine studies analyzing point-prevalence
rates for BED, N¼10 363 persons were screened at
first stage. A weighed mean prevalence rate of 3.53%
(95% CI [1.60, 6.13]) was identified (Fig. 4). Hetero-
geneity was large (I
2
¼97.00). Regarding potential
sources of heterogeneity, one study was identified
assessing the occurrence of two binges per week for
the shorter time range of 3 instead of 6 months as
proposed by the DSM-IV criteria for this disorder [33].
However, exclusion of this study did not reduce
heterogeneity meaningfully. Possible moderator vari-
ables for heterogeneity were age and sex, as studies
differed notably regarding age ranges. Due to the
restricted number of studies included, no moderator
analysis was conducted. Regarding country differ-
ences, no conclusion could be drawn as the preva-
lence rates varied strongly depending on the study.
DISCUSSION
The present systematic review and meta-analysis
evaluated the epidemiology of anorexia nervosa,
bulimia nervosa, and BED in Latin America, by
assessing not only English literature published
in international journals, but also by considering
high-quality studies published in Spanish or Portu-
guese in Latin American journals. We computed
mean point-prevalence rates for the specific
disorders and compared the prevalence rates of
the different countries.
Study
Angel 2008 0.83 0.53
4.55
0.00
0.00
0.38
0.53
0.64
0.00
1.20
0.95
2.83
0.85
0.14
0.27
0.55
1.21
7.93
0.36
0.22
2.28
1.96
1.43
0.64
2.76
2.27
8.58
3.78
2.89
2.00
1.98
13.14
8.29
7.58
12.64
3.39
4.43
12.17
4.59
7.17
7.77
5.17
4.99
5.27
3.40
100.00
6.13
0.05
0.00
1.15
1.14
1.00
0.21
1.90
1.54
5.35
2.08
1.15
0.95
1.16
Baader 2014
Compte 2015
Herscovici 2005
Mancilla-Diaz 2007a
Mancilla-Diaz 2007b
Mascarenhas 2007
Piñeros-Ortiz 2010
Pivetta 2010
Quintero-Párraga 2003
Rueda-Jaimes 2005a
Rueda-Jaimes 2005b
Rueda-Jaimes 2008
Vázquez 2005
Overall
Prev LCL UCL We igh t %
012345
Prevalence
67
FIGURE 3. Forest plot of the meta-analysis of the bulimia nervosa point-prevalence rates. LCL, lower confidence level; Prev,
prevalence rate; UCL, upper confidence level.
Study Prev LCL UCL Weight %
Baader 2014 2.55
0.00
9.78
6.60
5.03
3.22
1.82
0.66
4.21
1.55
0.00
8.23
5.54
4.21
1.83
1.13
0.29
2.07
3.79
0.36
11.46
7.74
5.92
4.97
2.66
1.17
7.04
10.44
9.18
14.70
16.89
16.46
6.73
9.31
10.16
6.13
Compte 2015
Freitas 2008
Herscovici 2005
Mascarenhas 2007
Morales 2015
Pivetta 2010
Quintero-Párraga 2003
Rueda-Jaimes 2008
Overall 3.53 1.60 6.13 100.00
012345
Prevalence
678910
FIGURE 4. Forest plot of the meta-analysis of the binge-eating disorder point-prevalence rates. LCL, lower confidence level;
Prev, prevalence rate; UCL, upper confidence level.
Eating disorders
368 www.co-psychiatry.com Volume 29 Number 6 November 2016
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Epidemiology of anorexia nervosa in Latin
America
Our findings demonstrate that the prevalence rates
for anorexia nervosa are comparable among the
different countries in Latin America, and signifi-
cantly lower compared to European or North Amer-
ican samples [4,6–9]. Similar prevalence rates were
found in Hispanic immigrants in the United States
[43]. This might be due to the different body ideal of
Latinas and Latinos compared to other ethnicities,
which idealizes a ‘curvier’ shape and higher weight
of the body than in Western countries [44
&&
,45,46],
and might be a protective factor in adolescence. As
recent research highlights the influence of other
etiological factors besides culture contributing to
the development and maintenance of anorexia
nervosa, genetic variations [47], or emotion dysre-
gulation [48] might play an important part in the
observed cases of anorexia nervosa in Latin America.
Epidemiology of bulimia nervosa in Latin
America
The point-prevalence of bulimia nervosa is mostly
reported as 1% in two-stage studies of young females
[7]. In this meta-analysis, the heterogeneity of the
studies was large. However, most of the studies
found prevalence rates of at least the same range
or higher than in Western Europe, but assessed
eating disorders of both genders. Additionally, half
of the samples were younger than 20 years of age. As
the peak incidence rate of bulimia nervosa is slightly
later than that of anorexia nervosa and females are
at a higher risk for developing any eating disorder,
this indicates that the general point-prevalence rate
in Latin America is most likely underestimated in
our study. This is in line with cross-cultural studies
conducted in the United States, indicating that the
prevalence rate of bulimia nervosa is at least in the
same range or even slightly higher in Hispanic
immigrants than in Caucasian residents [43,49].
Epidemiology of binge-eating disorder in
Latin America
We identified a higher BED point-prevalence in Latin
America than previous studies in Western popu-
lations [50,51]. Recent studies comparing eating
disorder prevalence rates of ethnic minorities with
Caucasian habitants identified higher prevalence
rates of Hispanic females in the United States
[43,49,52]. A possible explanation might be that food
has a high emotional value in many Latin American
cultures which reflects in the language (e.g., ‘Las
penas con pan duelen menos’, – the sorrows with
bread hurt less, ‘barriga llena, corazo
´n contento’ – a
full belly is a pleased heart), and addressing the
cultural assumptions regarding food is of importance
in nutritional interventions for Latinas [53
&&
].
Additionally, BED occurs more often in overweight
and obese individuals [54,55]. As most of the Latin
American countries have higher prevalence rates of
obesity and overweight than Western European
countries especially in females [56], a high prevalence
of BED might be a concomitant phenomenon of the
obesity epidemic in these countries.
Limitations of the study
One of the main difficulties in reviewing the liter-
ature regarding Latin American studies on eating
disorders is the wide range of instruments used to
assess the prevalence rates. The studies included
varied heavily regarding the study design (self-
administered questionnaire to two-phase designs)
and the instruments used. Hence, heterogeneity was
large and the mean prevalence rates of the meta-
analyses remain difficult to interpret. Furthermore,
the studies included were conducted during a time
span of 13 years. Because only few studies were from
Argentina, Chile, and Venezuela, we did not con-
duct cultural comparisons within Latin America and
the influence of specific countries on the prevalence
rates remains unexplored.
CONCLUSION
Eating disorders are common in Latin America, with
a lower prevalence rate for anorexia nervosa and a
higher rate for bulimia nervosa and BED, when
compared to Western Europe or the United States.
These findings might reflect a stronger cultural bond
of anorexia nervosa to a Western lifestyle than of
bulimia nervosa and BED. However, we could not
analyze cultural differences within Latin America as
most of the studies included were conducted either
in Brazil, Mexico, or Colombia. Therefore, further
research assessing the epidemiology and cultural
factors of eating disorders in developing countries
such as Ecuador, Suriname, or Guyana are needed to
draw conclusions regarding the cultural factors of
eating disorders in Latin America.
Acknowledgements
We are thankful to Jochem Ko
¨nig and Ekkehart Jenetzky
for their valuable comments on the methodology of this
study. Patricia Meinhardt conducted proof-reading of the
manuscript.
Financial support and sponsorship
None.
Epidemiology of eating disorders in Latin America Kolar et al.
0951-7367 Copyright ß2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com 369
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED
READING
Papers of particular interest, published within the annual period of review, have
been highlighted as:
&of special interest
&& of outstanding interest
1. Prince R. The concept of culture-bound syndromes: anorexia nervosa and
brain-fag. Soc Sci Med 1985; 21:197– 203.
2. Pike KM, Hoek HW, Dunne PE. Cultural trends and eating disorders. Curr
Opin Psychiatry 2014; 27:436–442.
3. Hoek HW. Epidemiology of eating disorders in persons other than the high-
risk group of young Western females. Curr Opin Psychiatry 2014; 27:423 –
425.
4. Smink FR, van Hoeken D, Donker GA, et al. Three decades of eating disorders
in Dutch primary care: decreasing incidence of bulimia nervosa but not of
anorexia nervosa. Psychol Med 2016; 46:1189 –1196.
5. Nagl M, Jacobi C, Paul M, et al. Prevalence, incidence, and natural course of
anorexia and bulimia nervosa among adolescents and young adults. Eur Child
Adolesc Psychiatry 2016; 25:903–918.
6. Hammerle F, Huss M, Ernst V, Bu
¨rger A. Thinking dimensional: prevalence of
DSM-5 early adolescent full syndrome, partial and subthreshold eating
disorders in a cross-sectional survey in German schools. BMJ Open
2016; 6:e010843.
7. Hoek HW. Incidence, prevalence and mortality of anorexia nervosa and other
eating disorders. Curr Opin Psychiatry 2006; 19:389–394.
8. Smink FRE, van Hoeken D, Hoek HW. Epidemiology of eating disorders:
incidence, prevalence and mortality rates. Curr Psychiatry Rep 2012;
14:406– 414.
9. Smink FRE, van Hoeken D, Hoek HW. Epidemiology, course, and outcome of
eating disorders. Curr Opin Psychiatry 2013; 26:543–548.
10. Tong J, Miao S, Wang J, et al. A two-stage epidemiologic study on prevalence
of eating disorders in female university students in Wuhan, China. Soc
Psychiatry Psychiatr Epidemiol 2014; 49:499–505.
11. Pike KM, Borovoy A. The rise of eating disorders in Japan: issues of culture
and limitations of the model of ‘westernization’. Cult Med Psychiatry 2004;
28:493– 531.
12. Musaiger AO, Al-Mannai M, Tayyem R, et al. Risk of disordered eating
attitudes among adolescents in seven Arab countries by gender and obesity:
a cross-cultural study. Appetite 2013; 60:162 –167.
13.
&
Doris E, Shekriladze I, Javakhishvili N, et al. Is cultural change associated with
eating disorders? A systematic review of the literature. Eat Weight Disord
2015; 20:149– 160.
A systematic review assessing the effect of acculturation to Western culture on
eating disorders. The authors evaluate mixed findings from 25 studies, stating that
even higher or less acculturation may be a risk factor for eating disorders – in
conclusion, further cross-cultural research is needed.
14. Kessler RC, Berglund PA, Chiu WT, et al. The prevalence and correlates of
binge eating disorder in the World Health Organization World Mental Health
Surveys. Biol Psychiatry 2013; 73:904– 914.
15. Palavras MA, Kaio GH, Mari JdJ, Claudino AM. A review of Latin American
studies on binge eating disorder. Rev Bras Psiquiatr 2011; 33 (Suppl
1):S81– S108.
16. van Hoeken D, Veling W, Smink FRE, Hoek HW. The incidence of anorexia
nervosa in Netherlands Antilles immigrants in the Netherlands. Eur Eat Disord
Rev 2010; 18:399– 403.
17. Marques L, Alegria M, Becker AE, et al. Comparative prevalence, correlates of
impairment, and service utilization for eating disorders across U.S. Ethnic
Groups: implications for reducing ethnic disparities in healthcare access for
eating disorders. Int J Eat Disord 2011; 44:412–420.
18. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for
systematic reviews and meta-analyses: the PRISMA statement. J Clin Epi-
demiol 2009; 62:1006– 1012.
19. Elm Ev, Altman DG, Egger M, et al. The Strengthening the Reporting of
Observational Studies in Epidemiology (STROBE) statement: guidelines for
reporting observational studies. PLoS Med 2007; 4:e296.
20. Loney PL, Chambers LW, Bennett KJ, et al. Critical appraisal of the health
research literature: prevalence or incidence of a health problem. Chronic Dis
Can 1998; 19:170– 176.
21. Doi SAR, Barendregt JJ, Khan S, et al. Advances in the meta-analysis of
heterogeneous clinical trials II: The quality effects model. Contemp Clin Trials
2015; 45 (Pt A):123–129.
22. Barendregt JJ, Doi SA, Lee YY, et al. Meta-analysis of prevalence. J Epidemiol
Community Health 2013; 67:974– 978.
23. A
´ngel LA, Martı
´nez LM, Go
´mez MT. Prevalencia de trastornos del comporta-
miento alimentario (TCA) en estudiantes de bachillerato [Prevalence of eating
disorders in high school students]. Rev Fac Med 2008; 56:193 –210.
24. Baader MT, Rojas CC, Molina FJL, et al. Diagno
´stico de la prevalencia de
trastornos de la salud mental en estudiantes universitarios y los factores de
riesgo emocionales asociados [Diagnostic of the prevalence of mental health
disorders in college students and associated emotional risk factors]. Rev Chil
Neuro-Psiquiatr 2014; 52:167–176.
25. Benjet C, Me
´ndez E, Borges G, Medina-Mora ME. Epidemiologı
´adelos
trastornos de la conducta alimentaria en una muestra representativa de
adolescentes [Epidemiology of eating disorders in a representative sample
of adolescents]. Salud Mental 2012; 35:483 –490.
26. Benjet C, Borges G, Mendez E, et al. Eight-year incidence of psychiatric
disorders and service use from adolescence to early adulthood: longitudinal
follow-up of the Mexican Adolescent Mental Health Survey. Eur Child Adolesc
Psychiatry 2016; 25:163– 173.
27.
&&
Compte EJ, Sepulveda AR, Torrente F. A two-stage epidemiological study of
eating disorders and muscle dysmorphia in male university students in
Buenos Aires. Int J Eat Disord 2015; 48:1092 –1101.
This study is the first study on eating disorders in young males in Argentina. They
found no case of anorexia nervosa, bulimia nervosa, or BED, but a 1.9% prevalenc e
of eating disorders not otherwise specified.
28. de Freitas SR, Appolinario JC, Souza AdM, Sichieri R. Prevalence of binge
eating and associated factors in a Brazilian probability sample of midlife
women. Int J Eat Disord 2008; 41:471–478.
29. Herscovici CR, Bay L, Kovalskys I. Prevalence of eating disorders in Argentine
boys and girls, aged 10 to 19, who are in primary care. A two-stage
community-based survey. Eat Disord 2005; 13:467 –478.
30. Bay LB, Rausch Herscovici C, Kovalskys I, et al. Alteraciones alimentarias en
nin
˜os y adolescentes argentinos que concurren al consultorio del pediatra
[Eating abnormalities in Argentinian children and adolescents occuring at
pediatric consultations]. Arch Argent Pediatr 2005; 103:305–316.
31. Swanson SA, Saito N, Borges G, et al. Change in binge eating and binge
eating disorder associated with migration from Mexico to the US. J Psychiatr
Res 2012; 46:31– 37.
32. Mancilla-Diaz JM, Franco-Paredes K, Vazquez-Arevalo R, et al. A two-stage
epidemiologic study on prevalence of eating disorders in female university
students in Mexico. Eur Eat Disord Rev 2007; 15:463– 470.
33. Lima Mascarenhas MT, Guimara
˜es de Almeida MM, Arau
´jo TMd, Kalil Prisco
AP. Transtornos alimentares na populac¸a
˜o de 20 a 59 anos de Feira de
Santana (BA), 2007 [Eating disorders in population aged 20 – 59 years from
Feira de Santana (BA), Brazil, 2007]. Cad Sau
´de Colet 2011; 19:179– 186.
34. Kalil Prisco AP, Araujo TMd, Guimara
˜es de Almeida MM, Bernardes Santos
KO. Prevalence of eating disorders in urban workers in a city of the northeast
of Brazil. Cien Saude Colet 2013; 18:1109–1118.
35.
&
Morales A, Gomes A, Jimenez B, et al. Binge eating disorder: prevalence,
associated factors and obesity in university students. Rev Colomb Psiquiatr
2015; 44:177– 182.
In this recent study, the prevalence rate of BED in Venezuelan students was 3.2%.
It is the first study based on the DSM-5 criteria for BED in Venezuela.
36. Ortı
´z SP, Caro JM, Mesa CLd. Factores de riesgo de los trastornos de la
conducta alimentaria en jo
´venes escolarizados en Cundinamarca (Colombia).
Rev Colomb Psiquiatr 2010; 39:313– 328.
37. Pivetta LA, Gonc¸alves-Silva RMV. Compulsa
˜o alimentar e fatores associados
em adolescentes de Cuiaba
´, Mato Grosso, Brasil [Binge eating and asso-
ciated factors among teenagers in Cuiaba
´, Mato Grosso State, Brazil]. Cad
Sau
´de Pu
´blica 2010; 26:337– 346.
38. Quintero-Pa
´rraga E, Pe
´rez-Montiel AC, Montiel-Nava C, et al. Trastornos
de la conducta alimentaria. Prevalencia y caracterı
´sticas clı
´nicas en
adolescentes de la ciudad de Maracaibo, Estado Zulia, Venezuela [Eating
behavior disorders. Prevalence and clinical features in adolescents in
the city of Maracaibo, Zulia State, Venezuela]. Invest Clı
´n 2003;
44:179– 193.
39. Rueda-Jaimes GE, Cadena Afanador. Laura del Pilar, et al. Validacio
´ndela
encuesta de comportamiento alimentario en adolescentes escolarizadas de
Bucaramanga, Colombia. Rev Colomb Psiquiatr 2005; 34:375 –385.
40. Rueda Jaimes GE, Dı
´az Martı
´nez LA, Ortiz Barajas DP, et al. Validacio
´n del
cuestionario SCOFF para el cribado de los trastornos del comportamiento
alimentario en adolescentes escolarizadas [Validation of the scoff question-
naire for screening the eating behaviour disorders of adolescents in school].
Aten Primaria 2005; 35:89–94.
41. Rueda-Jaimes GE, Camacho Lopez PA, Rangel-Martinez-Villalba AM. Internal
consistency and validity of the BITE for the screening of bulimia nervosa in
university students, Colombia. Eat Weight Disord 2008; 13:e35– e39.
42. Va
´zquez Are
´valo R, Mancilla Dı
´az JM, Mateo Gonza
´lez C, Aguilar L. Trastornos
del comportamiento alimentario y factores de riesgo en una muestra incidental
de jo
´venes mexicanos [Eating disorders and risk factors in an incidental
sample of young Mexicans]. Rev Mex Psicol 2005; 22:53 –63.
43. Alegria M, Woo M, Cao Z, et al. Prevalence and correlates of eating disorders
in Latinos in the United States. Int J Eat Disord 2007; 40 (Suppl):S15 –21.
44.
&&
Schooler D, Daniels EA. I am not a skinny toothpick and proud of it’: Latina
adolescents’ ethnic identity and responses to mainstream media images.
Body Image 2014; 11:11– 18.
Latina girls expressions’ of ethnicity was connected to more positive expressions
relating to their body after viewing thin-ideal White women. This supports the
theory that the Latina identity and body image might serve as a protective factor for
eating disorder when exposed to thin-ideal media.
Eating disorders
370 www.co-psychiatry.com Volume 29 Number 6 November 2016
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
45. Warren CS, Gleaves DH, Rakhkovskaya LM. Score reliability and factor
similarity of the Sociocultural Attitudes Towards Appearance Question-
naire-3 (SATAQ-3) among four ethnic groups. J Eat Disord 2013; 1:14.
46. Rakhkovskaya LM, Warren CS. Ethnic identity, thin-ideal internalization, and
eating pathology in ethnically diverse college women. Body Image 2014;
11:438– 445.
47. Brandys MK, de Kovel CGF, Kas MJ, et al. Overview of genetic research in
anorexia nervosa: The past, the present and the future. Int J Eat Disord 2015;
48:814– 825.
48. Haynos AF, Fruzzetti AE. Anorexia nervosa as a disorder of emotion dysre-
gulation: evidence and treatment implications. Clin Psychol: Sci Pract 2011;
18:183– 202.
49. Marques L, Alegria M, Becker AE, et al. Comparative prevalence, correlates of
impairment, and service utilization for eating disorders across US ethnic
groups: Implications for reducing ethnic disparities in healthcare access
for eating disorders. Int J Eat Disord 2011; 44:412 –420.
50. Striegel-Moore RH, Franko DL. Epidemiology of binge eating disorder. Int J
Eat Disord 2003; 34 (Suppl):S19– S29.
51. Davis C. The epidemiology and genetics of binge eating disorder (BED). CNS
Spectrums 2015; 20:522– 529.
52. Jennings KM, Kelly-Weeder S, Wolfe BE. Binge eating among racial
minority groups in the United States: an integrative review. J Am Psychiatr
Nurses Assoc 2015; 21:117– 125.
53.
&&
Reyes-Rodriguez ML, Gulisano M, Silva Y, et al. Las penas con pan duelen
menos’: The role of food and culture in Latinas with disordered eating
behaviors. Appetite 2016; 100:102– 109.
In this recent qualitative study the topics discussed by Latinas with ED in nutrition
therapy sessions were analyzed. Latin culture values were discussed in all of the
sessions, which shows the importance of cultural aspects in treatment of eating
disoders.
54. Marcus MD, Wildes JE. Disordered eating in obese individuals. Curr Opin
Psychiatry 2014; 27:443– 447.
55. Bertoli S, Leone A, Ponissi V, et al. Prevalence of and risk factors
for binge eating behaviour in 6930 adults starting a weight
loss or maintenance programme. Public Health Nutr 2016; 19:
71–77.
56. Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence
of overweight and obesity in children and adults during 1980– 2013: a
systematic analysis for the Global Burden of Disease Study 2013. Lancet
2014; 384:766– 781.
Epidemiology of eating disorders in Latin America Kolar et al.
0951-7367 Copyright ß2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com 371